SYSTEMS AND METHODS FOR DETECTING INSURANCE CLAIM FRAUD BY USING IMAGE DATA VALIDATION

Info

Publication number: 20230342855
Type: Application
Filed: Mar 18, 2020
Publication Date: Oct 26, 2023
Inventors: Kenneth Jason Sanchez (San Francisco, CA), Theobolt N. Leung (San Francisco, CA), Holger Struppek (San Francisco, CA), Scott Howard (San Francisco, CA), John Minichiello (Normal, IL), Vinay Kumar (Fremont, CA)
Application Number: 16/823,099

Abstract

Method and system for detecting potential fraudulent activities in vehicle insurance claims. For example, the method includes receiving, from a mobile computing device, a claim request comprising claims data, receiving, from the mobile computing device, a digital file comprising first images of a vehicle involved in an accident supporting the claim request, extracting metadata from the first images, comparing the metadata with the claims data, generating an assessment of the claim request based at least in part upon the comparing, and displaying the assessment of the claim request via a user interface.

Description

Description

CROSS-REFERENCE TO RELATED APPLICATIONS

This application claims priority to U.S. Provisional Patent Application No. 62/822,236, filed Mar. 22, 2019, incorporated by reference herein for all purposes.

FIELD OF THE DISCLOSURE

Some embodiments of the present disclosure are directed to detecting potential fraudulent activities in vehicle insurance claims. More particularly, certain embodiments of the present disclosure provide methods and systems for analyzing image data against claims data to generate an assessment of a claim request. Merely by way of example, the present disclosure has been applied to identifying errors and/or frauds in the image data of a vehicle submitted by a claimant. But it would be recognized that the present disclosure has much broader range of applicability.

BACKGROUND OF THE DISCLOSURE

Specific to the field of insurance is the need to accurately and efficiently verify facts related to claims and coverage. For example, if a holder of an automobile insurance policy is involved in a vehicle collision, a claims adjuster typically investigates the accident, determines who is at fault, and recommends how much an insurance company will pay for damages. Economies of scale and improved efficiencies can be obtained from not having to physically inspect the damaged property (e.g., automobiles or homes) when a claim arises. For automobile insurance, it is highly desirable to avoid physically sending a claims agent or adjuster to investigate the site of an accident or verify that the claim is legitimate.

BRIEF SUMMARY OF THE DISCLOSURE

Some embodiments of the present disclosure are directed to detecting potential fraudulent activities in vehicle insurance claims. More particularly, certain embodiments of the present disclosure provide methods and systems for analyzing image data against claims data to generate an assessment of a claim request. Merely by way of example, the present disclosure has been applied to identifying errors and/or frauds in the image data of a vehicle submitted by a claimant. But it would be recognized that the present disclosure has much broader range of applicability.

According to some embodiments, a method for detecting potential fraudulent activities in vehicle insurance claims include receiving a claim request from a mobile computing device via network communications where claim request includes claims data. Also, the method includes receiving a digital file from the mobile computing device via network communications where the digital file includes first images of a vehicle involved in an accident supporting the claim request. Additionally, the method includes extracting metadata from the first images and comparing the metadata with the claims data. Moreover, the method includes generating an assessment of the claim request based at least in part upon the comparing and displaying the assessment of the claim request via a user interface.

According to certain embodiments, a server for detecting potential fraudulent activities in vehicle insurance claims includes one or more processors and a memory storing instructions for execution by the one or more processors. The instructions, upon execution, cause the one or more processors to receive a claim request from a mobile computing device via network communications where the claim request includes claims data. Also, the instructions, upon execution, cause the one or more processors to receive a digital file from the mobile computing device via network communications where the digital file includes first images of a vehicle involved in an accident supporting the claim request. Additionally, the instructions, upon execution, cause the one or more processors to extract metadata from the first images and compare the metadata with the claims data. Moreover, the instructions, upon execution, cause the one or more processors to generate an assessment of the claim request based at least in part upon the comparing and display the assessment of the claim request via a user interface.

According to some embodiments, a non-transitory computer-readable medium stores instructions for detecting potential fraudulent activities in vehicle insurance claims. The instructions are executed by one or more processors of a computing device. The non-transitory computer-readable medium includes instructions to receive a claim request from a mobile computing device via network communications where the claim request includes claims data. Also, the non-transitory computer-readable medium includes instructions to receive a digital file from the mobile computing device via network communications where the digital file includes first images of a vehicle involved in an accident supporting the claim request. Additionally, the non-transitory computer-readable medium includes instructions to extract metadata from the first images and compare the metadata with the claims data. Moreover, the non-transitory computer-readable medium includes instructions to generate an assessment of the claim request based at least in part upon the comparing and display the assessment of the claim request via a user interface.

According to certain embodiments, a method for validating an insurance claim includes receiving a request to make a claim, such as a First Notice of Loss (FNOL) from a mobile device associated with a customer (e.g., a policyholder of auto insurance) via a network (e.g., Internet, mobile phone network, etc.). Also, the method includes receiving a digital file from the mobile device via the network. The digital file includes documents (e.g., police reports, policy information) and/or images (e.g., images taken of a damaged vehicle after an accident, images taken of a theft that took place inside a home, video recording of the vehicle after an accident). Additionally, the method includes validating the request by extracting metadata from the digital file and comparing the metadata with the request. Moreover, the method includes displaying results in response to comparing the metadata with the request.

According to some embodiments, a system for validating an insurance claim includes a mobile device associated with a customer. The mobile device is configured to transmit a request to make a claim and a digital file via a network. Also, the system includes a server configured to receive the digital file and the request via the network. The server is configured to validate the request by extracting metadata from the digital file and comparing the metadata with the request. The server is further configured to display results in response to comparing the metadata with the request.

Depending upon the embodiment, one or more benefits may be achieved. These benefits and various additional objects, features and advantages of the present disclosure can be fully appreciated with reference to the detailed description and accompanying drawings that follow.

BRIEF DESCRIPTION OF DRAWINGS

FIG. 1 is a simplified system for detecting potential fraudulent activities in vehicle insurance claims according to certain embodiments of the present disclosure.

FIG. 2 is a simplified mobile computing device for detecting potential fraudulent activities in vehicle insurance claims according to certain embodiments of the present disclosure.

FIG. 3 is a simplified server for detecting potential fraudulent activities in vehicle insurance claims according to certain embodiments of the present disclosure.

FIG. 4 is a simplified method for detecting one or more potential fraudulent activities in a vehicle insurance claim according to certain embodiments of the present disclosure.

FIG. 5 is a simplified method for detecting one or more potential fraudulent activities in a vehicle insurance claim according to some embodiments of the present disclosure.

DETAILED DESCRIPTION OF THE DISCLOSURE

Some embodiments of the present disclosure are directed to detecting potential fraudulent activities in vehicle insurance claims. More particularly, certain embodiments of the present disclosure provide methods and systems for analyzing image data against claims data to generate an assessment of a claim request. Merely by way of example, the present disclosure has been applied to identifying errors and/or frauds in the image data of a vehicle submitted by a claimant. But it would be recognized that the present disclosure has much broader range of applicability.

I. One or More Workflows for Detecting Potential Fraudulent Activities in Vehicle Insurance Claims According to Certain Embodiments

According to various embodiments, the present disclosure relates to using a fraud detection workflow to detect potential fraudulent activities in vehicle insurance claims and/or validate the vehicle insurance claims. In some embodiments, a server executes one or more algorithms, applications, programs, etc., to facilitate communication with a mobile computing device associated with a customer. For example, the customer prepares a request to make a claim (e.g., a FNOL) and a digital file using the mobile computing device. As an example, the digital file includes documents (e.g., police reports, policy information) and/or images (e.g., images taken of the type and/or extent of damage of a vehicle after an accident, images of the vehicle's VIN) supporting the claim request. In certain embodiments, a graphical user interface displayed via the mobile computing device guides or otherwise assists the customer in preparing the FNOL and the digital file. For example, the graphical user interface includes graphical elements (e.g., dropdowns, text boxes, icons, etc.) that are selectable by the customer to guide the customer in preparing the claim request and uploading particular documents and/or images. According to some embodiments, the server then receives, from the mobile computing device, the request to make the claim and the digital file. As an example, the server sends a confirmation message back to the mobile computing device indicating that the request and digital file was received successfully. According to certain embodiments, if additional data are needed to process the claim, the server provides a notification to the mobile computing device requesting the additional data. In some examples, the server instructs the customer of the mobile computing device to annotate one or more of the images with an indication of the vehicle damage. In certain examples, the customer uses the mobile computing device to provide the annotated images (e.g., circling the damage, highlighting the vehicle part having damage, etc.) back to the server. In some embodiments, the server transmits a notification to the mobile computing device requesting for images of the vehicle prior to the event (e.g., a vehicle accident) that gave rise to the claim.

In various embodiments, the server executes one or more algorithms, applications, programs, etc., to process the data in the fraud detection workflow. For example, the server extracts metadata from the images and compare the metadata with claims data included in the claim request. As an example, the server determines whether the images supporting the claim request accurately describe the incident reported in the claim request. For example, the server determines whether the date and/or location of the claim (e.g., the time and location of the vehicle accident) included in the claim request coincides with the date and/or location of the images (e.g., captured by the customer using the mobile computing device) as indicated by the metadata. As an example, any differences between the dates and/or locations are calculated and converted into a corresponding weight that is factored into an overall assessment of whether the claim is fraudulent. In some embodiments, various thresholds are considered. For example, if the metadata of an image indicates a date and/or location that is within a certain radius of the date and/or location of the vehicle accident as claimed in the claim request, then the server assigns a favorable weight to the image.

In some embodiments, if the server received images of the vehicle from the customer using the mobile computing device prior to the event (e.g., the vehicle accident) that gave rise to the claim, the server compares the images depicting the vehicle prior to the event with the images provided by the customer in support of the claim request. For example, the server classifies damages resulting from the event as well as any pre-existing damages to the vehicle that existed prior to the event. As an example, if the images provided by the customer in support of the claim request include pre-existing damages and the claim request seeks to compensate for the pre-existing damages, then the server determines that the claim request includes fraudulent activity, and accordingly adjusts the overall assessment of the claim. According to certain embodiments, the server either automatically omits the pre-existing damages during the claim assessment or notifies the customer to provide images that exclude the pre-existing damages and modifies the claim request to exclude the pre-existing damages. For example, if the images provided by the customer in support of the claim request do not include any pre-existing damages, the server proceeds to extract metadata from the images and compare the metadata with the claims data included in the request.

In certain embodiments, the server facilitates user interaction with a claims specialist (CS). For example, the CS is an insurance employee trained to perform manual review of the claim request with respect to the digital file. As an example, the CS oversees the fraud detection workflow as described above. For example, the CS analyzes the images that have been uploaded by the customer. As an example, the server acts as a feedback mechanism for the customer. For example, if additional images are needed to approve the claim, the CS contacts the customer via the server to request the additional images. As an example, the CS also annotates the images and provides the annotated images back to the customer to guide the customer in the collection of the additional images. For example, if the CS believes that additional angles or sides of the vehicle are needed to assess damage of the vehicle, the CS points out the particular sides of the vehicle as an annotation.

II. One or More Systems for Detecting Potential Fraudulent Activities in Vehicle Insurance Claims According to Certain Embodiments

FIG. 1 is a simplified system for detecting potential fraudulent activities in vehicle insurance claims according to certain embodiments of the present disclosure. This diagram is merely an example, which should not unduly limit the scope of the claims. One of ordinary skill in the art would recognize many variations, alternatives, and modifications. The system 100 includes a network 102, a mobile computing device 106, a vehicle 111, a server 114, a user 118, and a CS terminal 119. Although the above has been shown using a selected group of components for the system, there can be many alternatives, modifications, and variations. For example, some of the components may be expanded and/or combined. Other components may be inserted to those noted above. Depending upon the embodiment, the arrangement of components may be interchanged with others replaced. According to some embodiments, the system 100 includes hardware and/or software applications, as well as various data communication channels for facilitating data communications between the various hardware and software components.

In some embodiments, the network 102 is configured as any suitable network to facilitate communications between the mobile computing device 106 and the server 114. For example, the network 102 includes any appropriate combination of wired and/or wireless communication networks, and/or data transmission. As an example, the network 102 includes one or more telecommunication networks and constitutes (i) nodes and/or (ii) links used for data and/or communication exchange between various nodes. For example, the network 102 includes a wireless telephony network (e.g., GSM, CDMA, LTE, etc.), a Wi-Fi network (e.g., based at least in part upon 802.11x standards), a BLUETOOTH network, one or more proprietary networks, one or more base stations, access points, cellular networks, a secure public internet, a mobile-based network, a virtual private network or some other type of network, such as dedicated access lines, plain ordinary telephone lines, satellite links, a public switched telephone network (PSTN), etc., or any suitable combination thereof. In certain embodiments, the network 102 facilitates a connection to the Internet for the mobile computing device 106 and/or the server 114.

In certain embodiments, the mobile computing device 106 is permanently or removably installed in the vehicle 111 (e.g., a car, a truck, a motorcycle, etc.). For example, the mobile computing device 106 is permanently or removably installed in any suitable type of vehicle, such as an automobile, a watercraft, a motorcycle, etc.

In some embodiments, the mobile computing device 106 is configured to communicate with one or more of the network 102 and/or the vehicle 111 via any number of wired and/or wireless links. For example, the mobile computing device 106 communicates using a link 112a, a link 112b, and/or a link 112c via one or more suitable communication protocols that are the same communication protocols or different communication protocols. As an example, the mobile computing device 106 communicates with vehicle 111 via a BLUETOOTH communication protocol using the link 112b, while communicating with the server 114 via a cellular communication protocol using the network 102 and/or the link 112a and/or the link 112c.

According to certain embodiments, the mobile computing device 106 is implemented as any suitable type of computing device configured to perform the various functions as described herein, such as a smartphone, a tablet computer, a laptop computer, a wearable computing device, smart watch, smart glasses, personal digital assistant (PDA), or any suitable type of mobile device, including computing devices configured for wireless communication and data transmission. For example, the mobile computing device 106 is configured to execute one or more algorithms, program applications, etc., to receive one or more notifications inviting the customer to upload images showing the pre-existing condition of the vehicle 111, to submit a request for a claim, to capture images of the vehicle 111, to make annotations on the images of the vehicle 111, to provide documents (e.g., police reports), to submit a digital file encompassing the images and documents to the server 114 to support the claim, to receive one or more notifications that the damage of the vehicle has been approved or rejected based at least in part upon the claim request and digital file, to receive one or more notifications from the server 114, to view information associated with the customer's insurance policy profile stored on the server 114 and accessed via communications between the mobile computing device 106 and the server 114 utilizing the network 102, etc.

According to some embodiments, notifications are generated by the server 114 in accordance with any suitable schedule and/or upon any suitable number and type of conditions being satisfied. For example, a notification inviting the customer to upload images showing the pre-existing condition of the vehicle 111 is presented to the customer when the customer opens a mobile application installed in the mobile computing device 106 for the first time. As an example, a notification indicating that the damage of the vehicle has been approved or rejected is presented to the customer based at least in part upon whether a comparison of the claim request and digital file indicates fraudulent activity.

In certain embodiments, the mobile computing device 106 is configured to add metadata to images of the vehicle 111 captured via a camera associated with the mobile computing device 106. For example, the images are of vehicle 111 prior to an accident and/or after an accident. As an example, the images of the vehicle 111 prior to an accident depict any pre-existing damages prior to the accident. For example, the metadata include any suitable type of data, such as the current date, time, location, information regarding the environment, the weather conditions, etc. As an example, the mobile computing device 106 is configured to add metadata to the image prior to storing the images on the mobile computing device 106 and/or prior to sending the image of the vehicle 111 to the server 114.

In some embodiments, the mobile computing device 106 is configured to store data locally, such as images of the vehicle 111 showing any pre-existing damages, claims data associated with the claim request, the digital file including metadata associated with images of the vehicle 111 taken after an accident, a location of the mobile computing device 106, policy holder information, login credentials for a policy holder to access his/her policy profile via the server 114, etc. For example, the mobile computing device 106 stores user login and password data for a policy holder and/or one or more images of the vehicle 111.

In certain embodiments, the mobile computing device 106 sends the claims data and digital file to the server 114. For example, the server 114 assesses the claims data and digital file to determine whether the digital file supports the claim data.

In some embodiments, the mobile computing device 106 sends images of the vehicle 111 showing any pre-existing damages, the claims data, and digital file to the server 114. For example, the server 114 first assesses images of the vehicle 111 showing any pre-existing damages with respect to the digital file, and subsequently compares the claims data and digital file to determine whether the digital file supports the claim data.

In certain embodiments, if the digital file does not support the claim data, then the server 114 generates a manual review flag and appends or otherwise adds this data to the digital file and/or claims data. As an example, the review flag is an indicator for a CS to analyze the images that have been uploaded by the customer. For example, if additional images are needed to approve the claim, the CS contacts the customer to request for additional images. As an example, the server 114 provides such a request to the mobile computing device 106 via the network 102.

According to some embodiments, the server 114 is implemented as any suitable type of device to facilitate the various functions as described herein. For example, the server 114 is configured to execute various software applications, algorithms, and/or suitable programs facilitate one or more functions associated with the system 100. As an example, the server 114 is implemented as a network server, a web-server, a database server, a file server, a personal computer, a laptop, a smartphone, other mobile device, one or more database and/or storage devices, a shared database structure (e.g., comprising data storage using cloud computing technology), or any suitable combination thereof. For example, the server 114 is an insurance provider remote server.

According to certain embodiments, the server 114 is configured to communicate with the mobile computing device 106, the vehicle 111, and/or the CS terminal 119 (e.g., a desktop workstation accessible by the CS) via any suitable number and/or type of communication protocols. For example, communications between or among the mobile computing device 106, the server 114, the vehicle 111, and/or the CS terminal 119 result in any suitable type of data being transferred between any suitable combination of these devices, and implemented with any suitable number and/or type of wired and/or wireless links (e.g., the link 112a, the link 112b, the link 112c, and the link 112d).

In some embodiments, the mobile computing device 106 is configured to perform any suitable portion of processing functions remotely that are outsourced by the server 114. For example, the mobile computing device 106 assesses the claims data and digital file to determine whether the digital file supports the claim data.

In certain embodiments, the server 114 performs one or more actions when a manual review flag has been generated. For example, the server 114 sends a notification to the mobile computing device 106 that the policy profile associated with the customer has not been updated with images of the vehicle 111 showing any pre-existing damages, that a claim request has failed verification, that a claim request is potentially fraudulent, that manual review of the digital file is needed, that the customer should call a CS to manually validate the claim request, etc. As an example, if a received image has been flagged for manual review (e.g., the image is potentially fraudulent, the damage cannot be verified due to image quality, etc.), the server 114 transmits the flagged image (e.g., the image and a separate manual flag indication) to the CS terminal 119 for the CS to perform review.

In various embodiments, regardless of how the claim request and/or digital file are validated, the server 114 generates validation results based at least in part upon comparisons between images depicting the vehicle prior to the event with the images provided by the customer in support of the claim request, and/or between claims data and the digital file. For example, comparison results are converted to various pre-defined weights that adjust the overall assessment of the claim.

According to some embodiments, the server 114 determines whether the date and/or location of the claim (e.g., the time and location of a vehicle accident) included in the claim request coincides with the date and/or location of the image (e.g., captured by the customer using the mobile computing device 106) as indicated by its metadata. For example, any differences between the dates and/or locations are calculated and converted into a corresponding weight that is factored into an overall assessment of whether the claim is fraudulent. As an example, various thresholds are considered. For example, if the metadata of an image indicates a date and/or location that is within a certain radius of the date and/or location of the vehicle accident as claimed in the claim request, the server 114 assigns a favorable weight to the image.

According to certain embodiments, if the server 114 received images of the vehicle 111 from the customer using the mobile computing device 106 prior to the event (e.g., a vehicle accident) that gave rise to the claim, the server 114 compares the images depicting the vehicle 111 prior to the event with the images provided by the customer in support of the claim request. For example, if the images provided by the customer in support of the claim request include pre-existing damage and the claim request specifically seeks to compensate for the pre-existing damage, the server 114 determines that the claim request includes fraudulent activity, and accordingly adjust the overall assessment of the claim. As an example, if the images provided by the customer in support of the claim request do not include any pre-existing damage, the server 114 proceeds to extract metadata from the images and compare the metadata with claims data included in the request. For example, the server 114 is configured to assign a favorable weight to a claim request if the customer provided images of the vehicle 111 prior to the event that gave rise to the claim, with the assumption that customers who do so are acting responsibly.

According to some embodiments, the overall assessment of the claim is at the CS terminal 119 and/or the mobile computing device 106. For example, overall assessment of the claim includes a heuristic “percentile scale of fraud” suggesting the likelihood (e.g., 90%) that image(s) are fraudulent. As an example, the server 114 processes approved claims and subsequently adjusts or updates automobile insurance policies, premiums, and/or discounts, etc. as needed.

According to various embodiments, the mobile computing device 106, the server 114, and/or the CS terminal 119 include any suitable number of computing devices. For example, the vehicle 11 includes any suitable number of vehicles. In some embodiments, each computing device includes one or more CPUs and is configured to operate independently of other computing devices. For example, computing devices operating as a group process requests from other computing devices individually (e.g., based at least in part upon their availability) and/or concurrently (e.g., parallel processing). As an example, computing devices operating as a group process requests from other computing devices in a prioritized and/or distributed manner. For example, an operation associated with processing a claim request is performed on one computing device while another operation associated with processing the same claim request or a different claim request is performed on another computing device.

III. One or More Mobile Computing Devices for Detecting Potential Fraudulent Activities in Vehicle Insurance Claims According to Certain Embodiments

FIG. 2 is a simplified mobile computing device for detecting potential fraudulent activities in vehicle insurance claims according to certain embodiments of the present disclosure. This diagram is merely an example, which should not unduly limit the scope of the claims. One of ordinary skill in the art would recognize many variations, alternatives, and modifications. The mobile computing device 200 includes a controller 240, a display 216, a communication unit 230, a graphics processing unit (GPU) 218, a location acquisition unit 220, a speaker/microphone 222, an image capture device 226, and a user interface 228. Although the above has been shown using a selected group of components for the system, there can be many alternatives, modifications, and variations. For example, some of the components may be expanded and/or combined. Other components may be inserted to those noted above. Depending upon the embodiment, the arrangement of components may be interchanged with others replaced. In some embodiments, the mobile computing device 200 is an implementation of the mobile computing device 106 as shown in FIG. 1.

According to certain embodiments, the controller 240 includes a program memory 202, one or more of a microprocessor (MP) 206, a random-access memory (RAM) 208, and an input/output (I/O) interface 210, each of which may be interconnected via an address/data bus 212.

According to some embodiments, program memory 202 is implemented as a non-transitory tangible computer readable media configured to store computer-readable instructions, that when executed by the controller 240, cause the controller 240 to perform various acts. For example, the program memory 202 includes an operating system (OS) 242, one or more software applications 244, and/or one or more software routines 252. As an example, the program memory 202 is configured to include other portions to store data that are read from and written to by the MP 206, such as the data storage 260.

According to certain embodiments, the program memory 202 and/or the RAM 208 are implemented as any suitable type of memory, such as non-transitory computer readable memories, semiconductor memories, magnetically readable memories, and/or optically readable memories. For example, the MP 206 is configured to execute one or more of the OS 242, the software applications 244, the software routines 252, and/or other software applications.

According to some embodiments, the OS 242 is implemented as any suitable operating system platform depending on the particular implementation of the mobile computing device 200. For example, the OS 242 is implemented as one of a plurality of mobile platforms such as the iOS®, Android™, Palm® webOS, Windows® Mobile/Phone, BlackBerry® OS, or Symbian® OS mobile technology platforms, developed by Apple Inc., Google Inc., Hewlett-Packard Company, Microsoft Corporation, Research in Motion (RIM), and Nokia, respectively.

According to certain embodiments, the data storage 260 stores data used in conjunction with one or more functions performed by the mobile computing device 200 to facilitate the interaction between the mobile computing device 200 and one or more other devices, such as between the mobile computing device 200 and one or more networks (e.g., the network 102), vehicles (e.g., the vehicle 111), one or more servers (e.g., the server 114), etc. For example, the controller 240 is configured to communicate with additional data storage mechanisms (e.g., one or more hard disk drives, optical storage drives, solid state storage devices, etc.) that reside within or are otherwise associated with the mobile computing device 200.

According to some embodiments, the data storage 260 stores data such as application data for the one or more software applications 244, routine data for the one or more software routines 252, user policy profile information such as user identification, a user password, user name, and/or a user id, vehicle identifications, and/or images for a particular vehicle.

According to certain embodiments, the display 216 is implemented as any suitable type of display and may facilitate user interaction with the mobile computing device 200 in conjunction with the user interface 228. For example, the display 216 is implemented as a capacitive touch screen display, a resistive touch screen display, etc. As an example, the display 216 is configured to work in conjunction with the controller 240 and/or the GPU 218 to display information, such as instructions for taking a picture of a vehicle and uploading a digital file to submit a claim, user policy profile information, a confirmation that a claim has been validated, auto insurance premium or discount information, other auto insurance information, coverages, deductibles, etc.

According to some embodiments, the image capture device 226 is implemented as a camera integrated as part of the mobile computing device 200, a peripheral camera, a webcam, a camera installed inside a vehicle, a camera installed outside a vehicle, etc. For example, when the image capture device 226 is implemented as a device external to the mobile computing device 200, the image capture device 226 is configured to communicate with the mobile computing device 200 to send captured images to the mobile computing device 200. As an example, the image capture device 226 uses one or more orientations, zoom levels, effects, etc., to capture photos of a vehicle.

According to certain embodiments, the communication unit 230 is configured to facilitate communications between the mobile computing device 200 and one or more other devices. For example, the communication unit 230 is configured to support any suitable number and/or type of communication protocols based upon a particular network and/or device with which the mobile communication device 200 is communicating. As an example, the communication unit 230 supports communications in accordance with communication protocols such as cellular communication protocols (e.g., GSM, CDMA, LTE), communication protocols utilizing Wi-Fi 802.11 standards, WiMAX, near field communication (NFC) standards (e.g., ISO/IEC 18092, standards provided by the NFC Forum), BLUETOOTH communication protocols, etc.

According to some embodiments, the communication unit 230 is configured to support separate or concurrent communications based at least in part upon the same type of communication protocol or different types of communication protocols. For example, the communication unit 230 is configured to facilitate communications between the mobile computing device 200 and a server (e.g., the server 114) via a cellular communications protocol while facilitating communications between the mobile computing device 200 and a vehicle (e.g., the vehicle 111) in accordance with a BLUETOOTH communication protocol.

According to certain embodiments, the communication unit 230 transmits any suitable type of data to one or more servers (e.g., the server 114). For example, the communication unit 230 transmits one or more portions of claims data and/or digital file including images of a vehicle (e.g., the vehicle 111) as separate (e.g., concurrent or subsequent) data transmissions. As an example, the communication unit 230 transmits the one or more images including one or more portions of metadata. For example, the communication unit 230 transmits insurance policy information such as a policy number, a time and date stamp, vehicle information (e.g., a make, model, and year of the vehicle), user profile information (e.g., login credentials, a user ID), etc.

According to some embodiments, the location acquisition unit 220 is configured to utilize any suitable number and/or types of communication protocols to facilitate the determination of a geographic location of the mobile computing device 200. For example, the location acquisition unit 220 communicates with one or more satellites and/or wireless transmitters to determine a location of the mobile computing device 200. As an example, the location acquisition unit 220 utilizes a satellite Global Positioning System (GPS), an Assisted Global Positioning System (A-GPS), or any other suitable global positioning protocol (e.g., the GLONASS system operated by the Russian government, the Galileo system operated by the European Union, etc.) to determine a geographic location of the mobile computing device 200.

According to certain embodiments, the location acquisition unit 220 periodically stores, as part of the metadata, one or more locations of the mobile computing device 200 in any suitable portion of the program memory 202 (e.g., the data storage 260, the server 114, etc.). For example, the location acquisition unit 220 tracks the location of the mobile computing device 200 over time and stores these locations as part of the metadata (e.g., stored with the one or more images or separate from the one or more images).

According to some embodiments, one of the one or more software applications 244 is a claims submission application 246, which is implemented as a series of machine-readable instructions for performing the various tasks as described herein. As an example, the claims submission application 246 cooperates with one or more other hardware or software portions of the mobile computing device 200 to facilitate these functions. For example, the claims submission application 246 works in conjunction with one or more of the routines 252 and/or the other software applications 244 to perform one or more functions as described herein. As an example, the claims submission application 246 includes instructions for performing tasks, such as receiving user policy profile information from a server (e.g., the server 114), displaying one or more notifications, reminders, and/or instructions to capture one or more images of a vehicle (e.g., the vehicle 111), storing one or more images of a vehicle to a suitable portion of the data storage 260 and/or to another device (e.g., the server 114), adding metadata to one or more captured and/or stored images, sending one or more images to a server, flagging potentially fraudulent images, generating a failed image validation indicator, displaying user and/or policy information, facilitating communications between the mobile computing device 200 and one or more other devices in conjunction with the communication unit 230, etc.

According to certain embodiments, one of the one or more software applications 244 is a web browser 248. For example, the web browser 248 is a native web browser application, such as Apple's Safari®, Google Android™ mobile web browser, Microsoft Internet Explorer® for Mobile, Opera Mobile™, etc. As an example, the web browser 248 is implemented as an embedded web browser.

According to various embodiments, the web browser 248 is implemented as a series of machine-readable instructions that, when executed by the MP 206, result in the mobile computing device 200 receiving, interpreting, and/or displaying web page content from a server. For example, this web page content is utilized in conjunction with the claims submission application 246 to perform one or more functions as described herein.

According to some embodiments, one of the one or more software routines 252 includes a data read/write routine 254. For example, the data read/write routine 254 includes instructions, that when executed by the MP 206, cause the MP 206 to collect, measure, sample, generate, and/or store various types of data used by the claims submission application 246, such as claims data and a digital file. As an example, this data is stored in any suitable portion of the mobile computing device 200 (e.g., in the data storage 260 and/or the RAM 208) and/or to a server, which is accessed by the claims submission application 246 as needed.

According to certain embodiments, the data read/write routine 254 includes instructions that, when executed by the MP 206, cause the MP 206 to store geographic location data generated by the location acquisition unit 220 in the data storage 260. For example, the data read/write routine 254 includes instructions to cause the MP 206 to store a log of a time and/or a geographic location of the mobile computing device 200 when each image was captured by the image capture device 226, which is added to the one or more captured images as embedded metadata.

According to some embodiments, once one or more images of a vehicle are captured, damage and undamaged parts of a vehicle are identified, and the metadata of the one or more images (e.g., as part of the digital file) and the claims data are compared as requested by the customer. For example, the mobile computing device 200 offloads this processing via communications with a server, which is further discussed with reference to FIG. 3. As an example, the mobile computing device 200 performs this processing locally.

According to certain embodiments, with local processing, the mobile computing device 200 includes one or more imaging processing routines 256. For example, the imaging processing routines 256 include instructions that, when executed by the controller 240, cause the mobile computing device 200 to perform feature extraction, adaptive recognition, annotation, etc. on one or more images in order to identify content (e.g., various undamaged and/or damaged parts of a vehicle) and metadata associated with the one or more images.

According to some embodiments, if the digital file supports the claims data (e.g., time, location, and/or damage detected in images of the vehicle is consistent with the time, location, and/or damage indicated in the claims data), then the claims submission application 246 generates a validation indication (e.g., stored as part of the image as metadata), notify the user (e.g., via the display 216) that the digital file and/or claim has been validated, display the approved claim, communicate the approved claim and/or digital file to a server so that the customer's profile is updated accordingly, etc. For example, validation of the claim does not need manual review (e.g., by a claims specialist). As an example, comparing metadata of a digital file with the claims data provides an efficient process to reliably validate a claim request.

According to certain embodiments, if the digital file does not support the claims data (e.g., time, location, and/or damage detected in images of the vehicle is not within a predefined threshold of the time, location, and/or damage indicated in the claims data), then the claims submission application 246 generates a failed validation indicator (e.g., stored as part of the image as metadata), notifies the customer (e.g., via the display 216) and/or communicates the failed validation indication to a server, which includes one or more manual review flags.

According to some embodiments, a failed validation indication acts as a manual review flag to indicate that the associated image is to be manually reviewed when sent to a server by insurance personnel (e.g., by a claims specialist). For example, with local processing, the failed validation indication acts as a trigger to reset one or more processes, such as prompting the customer to capture another image or annotate an image.

According to certain embodiments, the image validation routine 258 includes instructions to facilitate the detection of a fraudulent image utilizing various types of data. For example, upon detection of a potential fraudulent image, an indication of the potential fraudulent image such as a manual review flag, is stored as part of the metadata in any suitable portion of the mobile computing device 200 (e.g., in the data storage 260 and/or the RAM 208) and/or to a server (e.g., the server 114), which is accessed by the claims submission application 246 as needed. As an example, upon receipt of the potential fraudulent data indicator, the appropriate personnel (e.g., insurance employees) follows up as needed to further investigate.

According to some embodiments, the image validation routine 258 facilitates the detection of a fraudulent reporting of a vehicle accident. For example, the image validation routine 258 determines whether mobile computing device 200 was in communication with a vehicle (e.g., via a BLUETOOTH connection) when the image was captured via the image capture device 226. As an example, the image validation routine 258 provides valuable information for determining whether a customer was located within a certain proximity to a vehicle when the image was captured.

According to certain embodiments, the image validation routine 258 facilitates the storage of geographic location data. For example, the geographic location data include a timestamp when an image was captured and/or a correlated geographic location of the mobile computing device 200. As an example, the image validation routine 258 compares the geographic location data with the claims data to determine whether the geographic location data are within a certain threshold of the claims data. For example, if the claim indicates that the customer was involved in an accident in Chicago, IL at 2:15 pm and the geographic location data of an image indicates that the image was taken in Chicago, IL at 2:10 pm, the image validation routine 258 determines that the geographic location data supports the claims data. As an example, if the claim indicates that the customer was involved in an accident in Chicago, IL at 2:15 pm and the geographic location data of an image indicates that the image was taken in Springfield, IL at 8:50 am, the image validation routine 258 determines that the geographic location data does not support the claims data.

According to some embodiments, the image validation routine 258 includes instructions, that when executed by the controller 240, perform one or more image analyses to determine whether an image is authentic or has been modified. For example, the image validation routine 258 facilitates the execution of any suitable number and/or type of routines to identify whether a customer has tampered with the image. As an example, this includes any suitable image forensic techniques, edge detection techniques, image error level analyses, colored layer analyses, compression analyses, etc. For example, the image validation routine 258 includes instructions to facilitate the identification and/or flagging of an image as fraudulent when the result of an image analysis indicates that an image has been digitally modified (e.g., via graphics editing software).

According to certain embodiments, one or more of the software applications 244 and/or the software routines 252 reside in the program memory 202 as default applications that are bundled together with the OS of the mobile computing device 200. For example, the web browser 248 is part of the software applications 244 that are included with the OS 242 implemented by mobile computing device 200.

According to some embodiments, one or more of the software applications 244 and/or the software routines 252 are installed on the mobile computing device 200 as one or more downloads, such as an executable package installation file downloaded from a suitable application store via a connection to the Internet. For example, the claims submission application 246, the data read/write routine 254, the one or more imaging processing routines 256, and/or the image validation routine 258 are stored in suitable portions of the program memory 202 upon installation of a package file downloaded in such a manner. For example, package download files include downloads via the iTunes store, the Google Play Store, the Windows Phone Store, downloading a package installation file from another computing device, etc. As an example, once downloaded, the claims submission application 246 is installed on mobile computing device 200 as part of an installation package such that, upon installation of the claims submission application 246 on the mobile computing device 200, the data read/write routine 254, the one or more imaging processing routines 256, and/or the image validation routine 258 are also installed.

According to certain embodiments, the controller 240 includes any suitable number and/or type of program memories, microprocessors, and/or RAM. According to some embodiments, the I/O interface 210 includes any suitable number and/or types of I/O interfaces.

IV. One or More Servers for Detecting Potential Fraudulent Activities in Vehicle Insurance Claims According to Certain Embodiments

FIG. 3 is a simplified server for detecting potential fraudulent activities in vehicle insurance claims according to certain embodiments of the present disclosure. This diagram is merely an example, which should not unduly limit the scope of the claims. One of ordinary skill in the art would recognize many variations, alternatives, and modifications. The server 300 includes a central processing unit (CPU) 306, a communication unit 308, and a memory 312. Although the above has been shown using a selected group of components for the system, there can be many alternatives, modifications, and variations. For example, some of the components may be expanded and/or combined. Other components may be inserted to those noted above. Depending upon the embodiment, the arrangement of components may be interchanged with others replaced. In some embodiments, the server 300 is an implementation of the server 114 as shown in FIG. 1.

According to various embodiments, the server 300 performs substantially similar functions as the mobile computing device 200. For example, a user or customer takes an image of a vehicle with the mobile computing device 200 and sends the image to the server 300 along with a claim request. As an example, the server 300 receives the claim request, images, and any associated metadata, policy information, vehicle information, time and date stamps, etc., performs image processing and/or validating to determine whether the image is potentially fraudulent, flags the image for manual review if the images do not support the claim request, and/or adjusts or updates automobile insurance policies, premiums, and/or discounts, etc.

According to some embodiments, the CPU 306 is configured to communicate with the memory 312 to store data to and read data from the memory 312. As an example, the memory 312 has a structure substantially similar to the program memory 202 and, when the various modules stored in the memory 312 are executed by the CPU 306, the various modules provide substantially the same functionality as the various modules stored in the program memory 202 when executed by the MP 206.

According to certain embodiments, the memory 312 includes a fraud detection application 313 and one or more memory modules utilized by the fraud detection application 313 such as a notification generator 315, a data read/write module 316, and an image validation module 324. For example, the fraud detection application 313 works in conjunction with the one or more modules to perform one or more functions as described herein. As an example, the fraud detection application 313 includes instructions that, when executed by the CPU 306, facilitate the implementation of a web-based and/or mobile-based application. For example, this application is utilized in conjunction with the server 300 to maintain policy profile information up-to-date. As an example, one or more policy holders communicate with the server 300 (e.g., via their respective mobile computing devices 200) to login and view their respective policy profiles, which include insurer contact information, policy information, risk assessment information, a listing of insurance policies and/or insured vehicles, etc.

According to some embodiments, the insured profile data are stored and accessed via the server 300 as customers interact with the fraud detection application 313. For example, the communication unit 308 has a structure substantially similar to the communication unit 230 and provides substantially the same functionality as the communication unit 230.

According to certain embodiments, the communication unit 308 is configured to facilitate data communications between the server 300, one or more mobile computing devices (e.g., the mobile computing device 200), and/or one or more terminals (e.g., the CS terminal 119). For example, the communication unit 308 performs such communications directly or indirectly (e.g., via network communications). As an example, the communication unit 308 sends data to and receives data from a mobile computing device including one or more images, insurance policy data, policy holder data (e.g., login credentials identifying the policy holder), and/or metadata associated with the one or more images, etc. For example, the communication unit 308 sends or receives suitable data as previously discussed above, such as an indication that an image having a manual review flag associated therewith has been resolved (e.g., once manual review has been completed), the damage of a vehicle determined from the resolution of one or more manual reviews, etc.

According to some embodiments, the CPU 306 executes instructions stored in the notification generator module 315 to generate notifications to one or more policy holders via the communication unit 308. For example, the CPU 306 executes instructions stored in the notification generator module 315 to query for information related to one or more policy holders. As an example, the CPU 306 causes a notification message to be sent via the communication unit 308 to one or more mobile computing devices. For example, an insurance policy holder receives a push notification on his/her mobile computing device from his/her insurance policy carrier to capture an image for a vehicle and provide a claim request associated with a policy. As an example, upon receipt of the notification, the policy holder provides a digital file (e.g., including an image of a vehicle and other documents) back to the server 300. For example, upon receipt of the image, the server 300 validates the image with respect to the claim request. As an example, additional notifications are sent to the mobile computing device to request additional data (e.g., images of a different angle of the vehicle) or annotations of the images.

According to certain embodiments, the CPU 306 executes instructions stored in the data read/write module 316 to perform acts substantially similar to those performed by the MP 206 when executing instructions stored in the read/write routine 254. For example, the CPU 306 executes instructions stored in the data read/write module 316 to read data from and/or to write data to the server 300 via the communication unit 308.

According to some embodiments, the CPU 306 executes instructions stored in the imaging processing module 318 to perform acts substantially similar to those performed by the MP 206 when executing instructions stored in the imaging processing routines 256. For example, the imaging processing module 318 includes instructions that, when executed by the CPU 306, cause the server 300 to perform feature extraction, adaptive recognition, annotation, etc. on one or more images in order to identify content (e.g., various undamaged and/or damaged parts of a vehicle) and metadata associated with the one or more images.

According to certain embodiments, the CPU 306 executes instructions stored in the image validation module 324 to perform acts substantially similar to those performed by the MP 206 when executing instructions stored in the image validation routine 258. For example, the CPU 306 executes instructions stored in the image validation module 324 to identify a potentially fraudulent image. As an example, the image validation module 324 includes instructions that, when executed by the CPU 306, cause the CPU 306 to analyze metadata received from the mobile computing device. For example, the CPU 306 additionally analyzes the claims data received from the mobile computing device and compares this data to the metadata to validate the claim request.

According to some embodiments, the CPU 306 executes instructions stored in the image validation module 324 to determine whether a mobile computing device 200 was located in proximity to a vehicle when the image was captured based at least in part upon metadata included in the image and/or received separately from the image. For example, the CPU 306 executes instructions stored in the image validation module 324 to determine whether geographic location data included as metadata in the image matches claims data associated with the customer's claim request. As an example, the CPU 306 executes instructions stored in the image validation module 324 to determine whether the image has been graphically manipulated. For example, the CPU 306 executes instructions stored in the image validation module 324 to determine whether the image includes any pre-existing damages of the vehicle based at least in part upon a comparison of the images of the digital file to images showing the pre-existing condition of the vehicle provided by the customer at a point in time prior to submission of the claim request.

According to certain embodiments, if the image is not validated and potentially fraudulent, then the fraud detection application 313 generates a failed image validation indicator. For example, generation of the failed image validation indicators triggers one or more acts. As an example, a failed image validation indicator triggers the communication unit 308 to store a manual review flag (e.g., same data as the failed image validation indicator) as part of the customer's updated profile data via execution of the data read/write module 316 by the CPU 306. For example, the generation of the failed image validation indicator triggers the communication unit 308 to send a notification to one or more insurance employees indicating the image needs manual review via execution of the notification generator module 315 by the CPU 306. As an example, generation of the failed image validation indicator triggers the communication unit 308 to send a notification to the mobile computing device indicating the manual review status and/or an indication that the claim request has not been successfully updated via execution of the notification generator module 315 by the CPU 306.

According to some embodiments, the CPU 306 executes instructions stored in the claims module 326 to display an overall assessment of the claim at the CS terminal 119. For example, overall assessment of the claim includes a heuristic “percentile scale of fraud” suggesting the likelihood (e.g., 90%) that image(s) are fraudulent. As an example, several factors affect the percentile scale. For example, any differences between the dates and/or locations in claims data and the dates and/or locations in image data are calculated and converted into a corresponding weight that is factored into an overall assessment of whether the claim is fraudulent. For example, if the dates and/or locations in claims data and the dates and/or locations in image data are within a particular threshold (e.g., same date, within 0.5 miles), the percentile scale of fraud is decreased. As an example, if the customer provided images of a vehicle prior to the event that gave rise to the claim, with the assumption that customers who do so are acting responsibly, the percentile scale of fraud is decreased. For example, if the images provided by the customer in support of the claim request include any pre-existing damage, the percentile scale of fraud is decreased.

V. One or More Methods for Detecting Potential Fraudulent Activities in Vehicle Insurance Claims According to Certain Embodiments

FIG. 4 is a simplified method for detecting one or more potential fraudulent activities in a vehicle insurance claim according to certain embodiments of the present disclosure. This diagram is merely an example, which should not unduly limit the scope of the claims. One of ordinary skill in the art would recognize many variations, alternatives, and modifications. The method 400 includes process 402 for receiving a claim request, process 404 for receiving a digital file including images, process 406 for extracting metadata from the images, process 408 for comparing the metadata with claims data, process 410 for generating an overall assessment of the claim request, and process 412 for displaying the overall assessment of the claim request. Although the above has been shown using a selected group of processes for the method, there can be many alternatives, modifications, and variations. For example, some of the processes may be expanded and/or combined. Other processes may be inserted to those noted above. Depending upon the embodiment, the sequence of processes may be interchanged with others replaced. For example, some or all processes of the method are performed by a computing device or a processor directed by instructions stored in memory. As an example, some or all processes of the method are performed according to instructions stored in a non-transitory computer-readable medium.

According to various embodiments, the method 400 is implemented by any suitable computing device (e.g., the mobile computing device 106 or the server 114, as shown in FIG. 1). For example, the method 400 is performed by one or more processors, applications, and/or routines, such as any suitable portion of the controller 240, the software applications 244, and/or the software routines 252, as shown in FIG. 2. As an example, the method 400 is performed by one or more processors, applications, and/or routines, such as any suitable portion of the CPU 306, the communication unit 308, the fraud detection application 313, and/or one or more of the notification generator module 315, the data read/write module 316, the imaging processing module 318, and/or the image validation module 324, as shown in FIG. 3.

At the process 402, one or more processors receive the claim request comprising the claims data according to some embodiments. For example, a customer (e.g., a policy holder) opens the one or more software applications 244 on the mobile computing device 106 to provide the claim request after the customer has been involved in a vehicle accident.

At the process 404, the one or more processors receive a digital file that includes documents and/or the images of the vehicle involved in the vehicle accident to support the claim request according to certain embodiments. For example, when the customer provides a police report and/or images capturing damages to the vehicle via the one or more software applications 244 on the mobile computing device 106.

At the process 406, the one or more processors extract the metadata from the images according to some embodiments. For example, the extraction includes extracting the time and location of when and where the images were captured.

At the process 408, the one or more processors compare the metadata from the images with the claims data according to certain embodiments. For example, the comparison includes identifying any differences between the metadata and the claims data and converting the differences into a corresponding weight.

At the process 410, the one or more processors generate the overall assessment of the claim request based at least in part upon the comparison according to some embodiments. For example, the overall assessment is generated to designate fraudulent activity if the comparison exceeds a threshold (e.g., the difference between the time as indicated by the metadata and the time as indicated by the claims data exceeds 2 days). As an example, the overall assessment is generated to designate non-fraudulent activity if the comparison is below a threshold (e.g., the difference between the location as indicated by the metadata and the location as indicated by the claims data is within a 0.5 mile radius).

At the process 412, the one or more processors display the overall assessment of the claim request via a user interface according to certain embodiments. For example, displaying the overall assessment includes displaying as a heuristic “percentile scale of fraud” at the mobile computing device 106 and/or at the CS terminal 119.

FIG. 5 is a simplified method for detecting one or more potential fraudulent activities in a vehicle insurance claim according to some embodiments of the present disclosure. This diagram is merely an example, which should not unduly limit the scope of the claims. One of ordinary skill in the art would recognize many variations, alternatives, and modifications. The method 500 includes process 502 for receiving images of a vehicle prior to an event, process 504 for receiving a claim request, process 506 for receiving images of the vehicle supporting the claim request, process 508 for comparing the images of the vehicle prior to the event with the images of the vehicle support the claim request, process 510 for determining whether the images supporting the claim request include pre-existing damage, process 512 for determining that the claim request includes fraudulent activity, process 514 for generating an overall assessment of the claim request, process 516 for extracting metadata from the images of the vehicle supporting the claim request, process 518 for comparing the metadata with claims data, and process 524 for displaying the overall assessment of the claim request. Although the above has been shown using a selected group of processes for the method, there can be many alternatives, modifications, and variations. For example, some of the processes may be expanded and/or combined. Other processes may be inserted to those noted above. Depending upon the embodiment, the sequence of processes may be interchanged with others replaced. For example, some or all processes of the method are performed by a computing device or a processor directed by instructions stored in memory. As an example, some or all processes of the method are performed according to instructions stored in a non-transitory computer-readable medium.

According to various embodiments, the method 500 is implemented by any suitable computing device (e.g., the mobile computing device 106 or the server 114, as shown in FIG. 1). For example, the method 500 is performed by one or more processors, applications, and/or routines, such as any suitable portion of the controller 240, the software applications 244, and/or the software routines 252 as shown in FIG. 2. As an example, the method 500 is performed by one or more processors, applications, and/or routines, such as any suitable portion of the CPU 306, the communication unit 308, the fraud detection application 313, and/or the one or more of the notification generator module 315, the data read/write module 316, the imaging processing module 318, and/or the image validation module 324 as shown in FIG. 3.

At the process 502, one or more processors receive the images of the vehicle prior to the event, such as a vehicle accident according to some embodiments. For example, a customer (e.g., a policy holder) opens the one or more software applications 244 on the mobile computing device 106 to provide the images of the vehicle showing any pre-existing conditions or damages before the customer has been involved in the vehicle accident.

At the process 504, the one or more processors receive the claim request comprising the claims data according to certain embodiments. For example, the customer opens the one or more software applications 244 on the mobile computing device 106 to provide the claim request after the customer has been involved in the vehicle accident.

At the process 506, the one or more processors receive a digital file that includes documents and/or the images of the vehicle involved in the vehicle accident to support the claim request according to some embodiments. For example, the customer provides a police report and/or the images capturing damages to the vehicle via the one or more software applications 244 on the mobile computing device 106.

At the process 508, the one or more processors compare the images of the vehicle prior to the vehicle accident with the images supporting the claim request according to certain embodiments. For example, the comparison determines whether any pre-existing damages (e.g., as would be shown in the images of the vehicle prior to the vehicle accident) are claimed in the claim request and included in the images supporting the claim request. As an example, the comparison includes wrapping (e.g., projecting) the images supporting the claim request onto the images of the vehicle prior to the vehicle accident to determine if any damages overlap.

At the process 510, the one or more processors determine whether the images supporting the claim request include the pre-existing damages and whether the claim request seeks to compensate for the pre-existing damages according to some embodiments.

At the process 512, if the images supporting the claim request include the pre-existing damages and the claim request seeks to compensate for the pre-existing damages, the one or more processors determine that the claim request includes fraudulent activity according to certain embodiments. At the process 514, the one or more processors generate the overall assessment of the claim request based at least in part upon the determination according to some embodiments. At the process 524, the one or more processors display the overall assessment of the claim request via a user interface according to certain embodiments.

At the process 512, if the images supporting the claim request do not include the pre-existing damages and the claim request does not seek to compensate for the pre-existing damages, the one or more processors determine that the claim request does not include fraudulent activity according to some embodiments. At the process 516, the one or more processors extract the metadata from the images of the vehicle supporting the claim request according to certain embodiments. At the process 518, the one or more processors compare the metadata from the images supporting the claim request with the claims data according to some embodiments. At the process 524, the one or more processors generate the overall assessment of the claim request based at least in part upon the comparison according to certain embodiments.

VI. Examples of Certain Embodiments of the Present Disclosure

According to some embodiments, a method for detecting potential fraudulent activities in vehicle insurance claims include receiving a claim request from a mobile computing device via network communications where claim request includes claims data. Also, the method includes receiving a digital file from the mobile computing device via network communications where the digital file includes first images of a vehicle involved in an accident supporting the claim request. Additionally, the method includes extracting metadata from the first images and comparing the metadata with the claims data. Moreover, the method includes generating an assessment of the claim request based at least in part upon the comparing and displaying the assessment of the claim request via a user interface. For example, the method is implemented according to at least FIG. 4 and/or FIG. 5.

According to certain embodiments, a server for detecting potential fraudulent activities in vehicle insurance claims includes one or more processors and a memory storing instructions for execution by the one or more processors. The instructions, upon execution, cause the one or more processors to receive a claim request from a mobile computing device via network communications where the claim request includes claims data. Also, the instructions, upon execution, cause the one or more processors to receive a digital file from the mobile computing device via network communications where the digital file includes first images of a vehicle involved in an accident supporting the claim request. Additionally, the instructions, upon execution, cause the one or more processors to extract metadata from the first images and compare the metadata with the claims data. Moreover, the instructions, upon execution, cause the one or more processors to generate an assessment of the claim request based at least in part upon the comparing and display the assessment of the claim request via a user interface. For example, the server is implemented according to at least FIG. 1 and/or FIG. 3.

According to some embodiments, a non-transitory computer-readable medium stores instructions for detecting potential fraudulent activities in vehicle insurance claims. The instructions are executed by one or more processors of a computing device. The non-transitory computer-readable medium includes instructions to receive a claim request from a mobile computing device via network communications where the claim request includes claims data. Also, the non-transitory computer-readable medium includes instructions to receive a digital file from the mobile computing device via network communications where the digital file includes first images of a vehicle involved in an accident supporting the claim request. Additionally, the non-transitory computer-readable medium includes instructions to extract metadata from the first images and compare the metadata with the claims data. Moreover, the non-transitory computer-readable medium includes instructions to generate an assessment of the claim request based at least in part upon the comparing and display the assessment of the claim request via a user interface. For example, the non-transitory computer-readable medium is implemented according to at least FIG. 1, FIG. 2, FIG. 3, FIG. 4 and/or FIG. 5.

VII. Examples of Machine Learning According to Certain Embodiments

According to some embodiments, machine learning techniques have been developed that allow parametric or nonparametric statistical analysis of large quantities of data. Such machine learning techniques may be used to automatically identify relevant variables (e.g., variables having statistical significance or a sufficient degree of explanatory power) from data sets. This may include identifying relevant variables or estimating the effect of such variables that indicate actual observations in the data set. This may also include identifying latent variables not directly observed in the data, e.g., variables inferred from the observed data points. In some embodiments, the methods and systems described herein may use machine learning techniques to identify and estimate the effects of observed or latent variables such as vehicle location, time of day, type of vehicle collision, type of vehicle damage or personal injury, vehicle collision location, amount of vehicle damage or medical expenses associated with a vehicle collision, and/or other such variables that influence the risks associated with vehicle collisions or vehicle travel.

According to certain embodiments, although the methods described herein may not directly mention machine learning techniques, such methods may be read to include such machine learning for any determination or processing of data that may be accomplished using such techniques. In some embodiments, such machine-learning techniques may be implemented automatically upon occurrence of certain events or upon certain conditions being met. Use of machine learning techniques, as described herein, may begin with training a machine learning program (e.g., to detect damages of a vehicle based on hundreds or thousands of images), or such techniques may begin with a previously trained machine learning program. The machine learning programs may utilize deep learning algorithms that are primarily focused on pattern recognition and may be trained after processing multiple examples. The machine learning programs may include Bayesian Program Learning (BPL), image or object recognition, and/or optical character recognition.

According to some embodiments, a processor or a processing element may be trained using supervised machine learning and/or unsupervised machine learning, and the machine learning may employ an artificial neural network, which, for example, may be a convolutional neural network, a deep learning neural network, or a combined learning module or program that learns in two or more fields or areas of interest. Machine learning may involve identifying and recognizing patterns in existing data in order to facilitate making predictions for subsequent data. Models may be created based upon example inputs in order to make valid and reliable predictions for novel inputs.

VIII. Additional Considerations According to Certain Embodiments

For example, some or all components of various embodiments of the present disclosure each are, individually and/or in combination with at least another component, implemented using one or more software components, one or more hardware components, and/or one or more combinations of software and hardware components. As an example, some or all components of various embodiments of the present disclosure each are, individually and/or in combination with at least another component, implemented in one or more circuits, such as one or more analog circuits and/or one or more digital circuits. For example, while the embodiments described above refer to particular features, the scope of the present disclosure also includes embodiments having different combinations of features and embodiments that do not include all of the described features. As an example, various embodiments and/or examples of the present disclosure can be combined.

Additionally, the methods and systems described herein may be implemented on many different types of processing devices by program code comprising program instructions that are executable by the device processing subsystem. The software program instructions may include source code, object code, machine code, or any other stored data that is operable to cause a processing system to perform the methods and operations described herein. Certain implementations may also be used, however, such as firmware or even appropriately designed hardware configured to perform the methods and systems described herein.

The systems' and methods' data (e.g., associations, mappings, data input, data output, intermediate data results, final data results) may be stored and implemented in one or more different types of computer-implemented data stores, such as different types of storage devices and programming constructs (e.g., RAM, ROM, EEPROM, Flash memory, flat files, databases, programming data structures, programming variables, IF-THEN (or similar type) statement constructs, application programming interface). It is noted that data structures describe formats for use in organizing and storing data in databases, programs, memory, or other computer-readable media for use by a computer program.

The systems and methods may be provided on many different types of computer-readable media including computer storage mechanisms (e.g., CD-ROM, diskette, RAM, flash memory, computer's hard drive, DVD) that contain instructions (e.g., software) for use in execution by a processor to perform the methods' operations and implement the systems described herein. The computer components, software modules, functions, data stores and data structures described herein may be connected directly or indirectly to each other in order to allow the flow of data needed for their operations. It is also noted that a module or processor includes a unit of code that performs a software operation, and can be implemented for example as a subroutine unit of code, or as a software function unit of code, or as an object (as in an object-oriented paradigm), or as an applet, or in a computer script language, or as another type of computer code. The software components and/or functionality may be located on a single computer or distributed across multiple computers depending upon the situation at hand.

The computing system can include client devices and servers. A client device and server are generally remote from each other and typically interact through a communication network. The relationship of client device and server arises by virtue of computer programs running on the respective computers and having a client device-server relationship to each other.

This specification contains many specifics for particular embodiments. Certain features that are described in this specification in the context of separate embodiments can also be implemented in combination in a single embodiment. Conversely, various features that are described in the context of a single embodiment can also be implemented in multiple embodiments separately or in any suitable subcombination. Moreover, although features may be described above as acting in certain combinations, one or more features from a combination can in some cases be removed from the combination, and a combination may, for example, be directed to a subcombination or variation of a subcombination.

Similarly, while operations are depicted in the drawings in a particular order, this should not be understood as requiring that such operations be performed in the particular order shown or in sequential order, or that all illustrated operations be performed, to achieve desirable results. In certain circumstances, multitasking and parallel processing may be advantageous. Moreover, the separation of various system components in the embodiments described above should not be understood as requiring such separation in all embodiments, and it should be understood that the described program components and systems can generally be integrated together in a single software product or packaged into multiple software products.

Although specific embodiments of the present disclosure have been described, it will be understood by those of skill in the art that there are other embodiments that are equivalent to the described embodiments. Accordingly, it is to be understood that the present disclosure is not to be limited by the specific illustrated embodiments.

Claims

1. A computer-implemented method for detecting potential fraudulent activities, the method comprising:

receiving, by one or more processors, a submission from a mobile computing device via network communications, the submission comprising accident data, the accident data comprising an accident location indicating a location of an accident;

receiving, by the one or more processors, a digital file from the mobile computing device via network communications, the digital file comprising one or more first images of a vehicle involved in the accident supporting the submission;

extracting, by the one or more processors, metadata from the one or more first images, the metadata comprising an image location indicating a location where the one or more first images are captured;

determining whether the image location is within a predetermined radius of the accident location based on the extracted metadata;

generating, by the one or more processors, an assessment of the submission by at least: if the image location is determined to be within the predetermined radius of the accident location, assigning a first weight to the one or more first images; and if the image location is determined to be outside the predetermined radius of the accident location, assigning a second weight to the one or more first images, the second weight being different from the first weight; and generating the assessment of the submission using the assigned first weight or the assigned second weight;

analyzing, by the one or more processors, the one or more first images to determine, by performing a feature extraction or an adaptive recognition process, whether damage of the vehicle can be identified; and

transmitting, by the one or more processors, a notification to the mobile computing device via network communications to request for additional data if the damage of the vehicle cannot be identified.

2. The computer-implemented method of claim 1, further comprising:

receiving, by the one or more processors, one or more second images of the vehicle prior to the accident;

comparing, by the one or more processors, the one or more second images with the one or more first images;

detecting, by the one or more processors, pre-existing damage of the vehicle in the one or more first images based at least in part upon the comparing, wherein the pre-existing damage of the vehicle is designated to be compensated in the accident data; and

adjusting the assessment of the submission based at least in part upon the detecting.

3. (canceled)

4. (canceled)

5. The computer-implemented method of claim 1, wherein the additional data comprises one or more additional images of the damage of the vehicle at different angles.

6. The computer-implemented method of claim 1, wherein the additional data comprises annotations of the damage of the vehicle on the one or more first images.

7. The computer-implemented method of claim 1, further comprising:

generating, by the one or more processors, a failed validation indicator when the metadata and the accident data do not match.

8. The computer-implemented method of claim 1, further comprising:

generating, by the one or more processors, a manual review flag when the metadata and the accident data do not match.

9. A server for detecting potential fraudulent activities, the server comprising:

a memory storing instructions; and

one or more processors configured to, upon execution of the instructions: receive a submission from a mobile computing device via network communications, the submission comprising accident data, the accident data comprising an accident location indicating a location of an accident; receive a digital file from the mobile computing device via network communications, the digital file comprising one or more first images of a vehicle involved in the accident supporting the submission; extract metadata from the one or more first images, the metadata comprising an image location indicating a location where the one or more first images are captured; determine whether the image location is within a predetermined radius of the accident location based on the extracted metadata; generate an assessment of the submission by at least: if the image location is determined to be within the predetermined radius of the accident location, assigning a first weight to the one or more first images; and if the image location is determined to be outside the predetermined radius of the accident location, assigning a second weight to the one or more first images, the second weight being different from the first weight; and generating the assessment of the submission using the assigned first weight or the assigned second weight; analyze the one or more first images to determine, by performing a feature extraction or an adaptive recognition process, whether damage of the vehicle can be identified; and transmit a notification to the mobile computing device via network communications to request for additional data if the damage of the vehicle cannot be identified.

10. The server of claim 9, wherein the one or more processors is further configured to:

receive one or more second images of the vehicle prior to the accident;

compare the second images with the first images;

detect pre-existing damage of the vehicle in the first images based at least in part upon the comparing, wherein the pre-existing damage of the vehicle is designated to be compensated in the accident data; and

adjust the assessment of the submission based at least in part upon the detecting.

11. (canceled)

12. (canceled)

13. The server of claim 9, wherein the additional data comprises one or more additional images of the damage of the vehicle at different angles.

14. The server of claim 9, wherein the additional data comprises annotations of the damage of the vehicle on the one or more first images.

15. The server of claim 9, wherein the one or more processors is further configured to generate a failed validation indicator when the metadata and the accident data do not match.

16. The server of claim 9, wherein the one or more processors is further configured to generate a manual review flag when the metadata and the accident data do not match.

17. A non-transitory computer-readable medium storing instructions for detecting potential fraudulent activities that, when executed by one or more processors, cause the one or more processors to:

receive a submission from a mobile computing device via network communications, the submission comprising accident data, the accident data comprising an accident location indicating a location of an accident;

receive a digital file from the mobile computing device via network communications, the digital file comprising one or more first images of a vehicle involved in the accident supporting the submission;

extract metadata from the one or more first images, the metadata comprising an image location indicating a location where the one or more first images are captured;

determine whether the image location is within a predetermined radius of the accident location based on the extracted metadata;

generate an assessment of the submission by at least: if the image location is determined to be within the predetermined radius of the accident location, assigning a first weight to the one or more first images; and if the image location is determined to be outside the predetermined radius of the accident location, assigning a second weight to the one or more first images, the second weight being different from the first weight; and generating the assessment of the submission using the assigned first weight or the assigned second weight;

analyze the one or more first images to determine, by performing a feature extraction or an adaptive recognition process, whether damage of the vehicle can be identified; and

transmit a notification to the mobile computing device via network communications to request for additional data if the damage of the vehicle cannot be identified.

18. The non-transitory computer-readable medium of claim 17, wherein the instructions that, when executed by the one or more processors, further cause the one or more processors to:

receive one or more second images of the vehicle prior to the accident;

compare the one or more second images with the one or more first images;

detect pre-existing damage of the vehicle in the first images based at least in part upon the comparing, wherein the pre-existing damage of the vehicle is designated to be compensated in the accident data; and

adjust the assessment of the submission based at least in part upon the detecting.

19. (canceled)

20. (canceled)

21. The non-transitory computer-readable medium of claim 17, wherein the additional data comprises one or more additional images of the damage of the vehicle at different angles.

22. The non-transitory computer-readable medium of claim 17, wherein the additional data comprises annotations of the damage of the vehicle on the one or more first images.

23. The non-transitory computer-readable medium of claim 17, wherein the instructions that, when executed by the one or more processors, further cause the one or more processors to generate a failed validation indicator when the metadata and the accident data do not match.

24. The non-transitory computer-readable medium of claim 17, wherein the instructions that, when executed by the one or more processors, further cause the one or more processors to generate a manual review flag when the metadata and the accident data do not match.