Abstract: A driver monitoring system according to an example of the present disclosure includes a camera configured to record images of a driver within a cabin of a vehicle. At least one vehicle sensor is configured to detect an anomalous driving event of the vehicle based on predefined criteria. A controller is in communication with the camera and the at least one vehicle sensor. The controller is configured to, based on the detection, obtain a particular image taken by the camera that depicts the driver during the anomalous event. The controller is configured to transmit the particular image to a fleet manager, store the particular image in a local repository of anomalous driving images, or both.

Abstract: The method of matching digital images of the same article in a data processor unit comprises the steps of: transforming each digital image of an article into a local divergence topographic map of the luminance gradient vector field; detecting singularities or extrema of local divergence in the luminance gradient vector field, such singularities corresponding to points of interest in said digital image; and, for each detected point of interest, encoding the values for the singularity of the gradient field that are located on a plurality of concentric rings centered on the point of interest so as to derive a digital data vector (210); and transforming said vector into a digital hash key (220) by means of a family of hash functions of the cosine Locality-Sensitive Hashing (LSH) type.

Abstract: In some embodiments, a method inputs a set of images into a network and trains the network based on a classification of the set of images to one or more characters in a set of characters. The method obtains a set of encodings for the one or more characters based on a layer of the network that restricts the output of the layer to a number of values. Then, the method stores the set of encodings for the one or more characters, wherein an encoding in the set of encodings is retrievable when a corresponding character is determined.

Abstract: Described is a digitized mail control system. The system may include a server having a memory storing mail data associated with a user and a user computing device coupled to the server. The server may be programmed to receive an access signal from the user computing device that has accessed the system. In response to receiving the access signal, the server may process the user data and locate the stored mail data associated with the user data. The server may then create and send for execution on the user computing device instruction to display mail data including sender information and action buttons associated with predetermined actions to take in response to the sender information. The server may receive a signal including a selected action associated with a selected action button with regard to a selected sender and automatically execute program code corresponding to the selected action.

Abstract: A system and method for managing packages may include determining an identifier on a package. A determination as to whether the package is to be positioned on a pallet based on the identifier may be made. In response to determining that the package is not to be positioned on the pallet, a first feedback signal (e.g., red light) indicative to the operator that the package is not to be placed on the pallet may be generated. In response to determining that the package is to be positioned on the pallet, a second feedback signal (e.g., green light) that is indicative to the operator that the package is to be on the pallet may be generated or no signal may be generated. Additionally, if an identifier cannot be read from the package, then a third feedback signal (e.g., yellow light) may be generated and illuminated on the package or elsewhere.

Abstract: Systems and methods for applying data optimization to images to meet a computer vision requirement are disclosed. According to an aspect, a method includes determining a computer vision requirement of a remote computing device. The method also includes determining, based on the computer vision requirement, a data optimization technique to apply to multiple images. Further, the method includes applying the data optimization technique to one or more of the images to generate data for communication. The method also includes communicating the data to the remote computing device.

Abstract: Systems, methods, and computer program products for smart upload automation in which actions are automatically performed on a set of digital assets against a target item. In one embodiment, a system includes a network, a server machine, a client machine and a data storage device, each of which is coupled to the network. The client machine designates digital assets and a target item against which the assets will be uploaded. The digital assets are uploaded by the client machine to the data storage device via the network. The server machine automatically performs actions on the digital assets without intervention by the client machine, where the actions are associated with or in some way defined by the target item. The actions may include setting metadata values of the digital assets based upon metadata associated with the target item, or generating different renditions of the digital assets.

Abstract: Improving visual quality of a raster image includes detecting connectivity components, detecting defects in each of the connectivity components based on a characteristic line width thereof, detecting segments in each of the connectivity components, detecting joints based on geometry of the connectivity components, creating a structural graph based on the segments and joints, and correcting the raster image according to the structural graph and detected ones of the defects. The joints may correspond to linear joints, T-joints, or X-joints. Detecting types of joints may include determining a configuration of adjacent segments in a proximity of each of the joints. A characteristic line width may be determined by determining co-boundaries on opposite sides of each of the segments and determining average distances between the co-boundaries. The raster image may be a binary black-and-white image of a line drawing obtained from a photograph or a scan of a handwritten document.

Abstract: Embodiments herein provide computer-implemented techniques for allowing a user computing device to extract financial card information using optical character recognition (“OCR”). Extracting financial card information may be improved by applying various classifiers and other transformations to the image data. For example, applying a linear classifier to the image to determine digit locations before applying the OCR algorithm allows the user computing device to use less processing capacity to extract accurate card data. The OCR application may train a classifier to use the wear patterns of a card to improve OCR algorithm performance. The OCR application may apply a linear classifier and then a nonlinear classifier to improve the performance and the accuracy of the OCR algorithm. The OCR application uses the known digit patterns used by typical credit and debit cards to improve the accuracy of the OCR algorithm.

Abstract: The invention relates to a method of controlling a touch-surface control device characterized in that it comprises a step of shape recognition of a control trajectory on a touch surface from among at least two predefined shapes of trajectory, in which in the course of a predetermined duration (dT): the control trajectory is sampled in order to determine a sampled angle (d?) of the control trajectory for each sampling period (Te), a parameter representative of the evolution of at least two sampled angles (d?) is compared with a predetermined threshold, and a predefined trajectory shape is assigned to the control trajectory as a function of the result of the comparison. The invention also relates to a control device comprising a tough surface characterized in that it comprises a processing unit for implementing a control method as described above.

Abstract: In a case where a light distribution control apparatus judges that there is a possibility that a driver's own vehicle and an animal will contact each other if it is assumed that the animal remains at the present position, then it controls headlight such that the head of the animal is out of the illuminating range of the headlight. In a case where the light distribution control apparatus judges that there is no possibility that the driver's own vehicle and the animal will contact each other if it is assumed that the animal remains at the present position, then it controls the headlight such that the head of the animal is within the illuminating range of the headlight.

Abstract: A system, method, and computer readable medium for mail analysis. A method includes receiving in a data processing system an image of a first mailpiece and associated machine-recognized data. The method includes comparing the machine-recognized data with a target list to determine a match with a target individual. The method includes performing a writer identification process on the image of the first mailpiece when a match is determined between the machine-recognized data and the target individual. The writer identification process produces writer identification data associated with the first mailpiece. The method includes storing the image of the first mailpiece and associated machine-recognized data and writer identification data.

Abstract: A system, method and program product for providing a reflexive data collection system. A system is provided that includes: a system for defining a set of environmental conditions; a device for automatically implementing the set of environmental conditions; a system for collecting biometric data under the set of environmental conditions; and a system for encoding the biometric data and the set of environmental conditions into a reflexive template.

Abstract: Methods, systems and computer program products are described for providing a computer implemented control system for an automated system. The system includes a document receiving component configured for receiving a plurality of design documents and for rendering a plurality of design document views corresponding to the plurality of design documents. The system also includes an input receiving component configured for receiving an indication selecting first and second design document views, a data extraction component configured for automatically extracting design data from the first and second design document views, a matching component configured for determining a relationship between the design data from the first and second design document views, and a data manager component configured for storing the relationship in a data store accessible by a computer implemented control system that controls a plurality of devices in the automated system based on the stored relationship.

Abstract: A nontransitory computer readable medium having stored thereon software instructions that, when executed by a computer, cause the computer to perform operations. The operations can include intercepting a request for manual processing of a mail piece image. The request can be transmitted by an originating MPE having failed to resolve the mail piece image. The request can be ignored and the mail piece scan image can be forwarded to an automated mail recognition system (AMRS). If the AMRS successfully resolved the image, a response, including a plurality of addressing data corresponding to the mail piece scan image, can be received from the AMRS. If the AMRS successfully resolved the image, at least a portion of the plurality of addressing data can be transmitted to the originating MPE. If the AMRS did not successfully resolve the image, the request can be forwarded to a manual mail recognition processing system.

Abstract: A system, apparatus, and method for detecting overlapped items in a sequence of items moving along a conveying path based on rigidity and thickness contour measurements.

Abstract: Techniques may comprise identifying surfaces, textures, and object dimensions from unorganized point clouds derived from a capture device, such as a depth sensing device. Employing target digitization may comprise surface extraction, identifying points in a point cloud, labeling surfaces, computing object properties, tracking changes in object properties over time, and increasing confidence in the object boundaries and identity as additional frames are captured. If the point cloud data includes an object, a model of the object may be generated. Feedback of the model associated with a particular object may be generated and provided real time to the user. Further, the model of the object may be tracked in response to any movement of the object in the physical space such that the model may be adjusted to mimic changes or movement of the object, or increase the fidelity of the target's characteristics.

Abstract: A nontransitory computer readable medium having stored thereon software instructions that, when executed by a computer, cause the computer to perform operations. The operations can include intercepting a request for manual processing of a mail piece image. The request can be transmitted by an originating MPE having failed to resolve the mail piece image. The request can be ignored and the mail piece scan image can be forwarded to an automated mail recognition system (AMRS). If the AMRS successfully resolved the image, a response, including a plurality of addressing data corresponding to the mail piece scan image, can be received from the AMRS. If the AMRS successfully resolved the image, at least a portion of the plurality of addressing data can be transmitted to the originating MPE. If the AMRS did not successfully resolve the image, the request can be forwarded to a manual mail recognition processing system.

Abstract: A computerized method for retrofitting a legacy manual mail recognition processing system (MRPS) with an automated MRPS without substantial changes to the configuration of the legacy manual MRPS includes detecting a request by a keyer desk software component to display a mail piece scan image on a computer display connected to the keyer desk computer. The image can at least temporarily be prevented from being displayed on the computer display and the image can be read from memory. Automated recognition processing can be performed on the image to resolve the image and create an automated recognition processing results data. If said automated recognition processing was successful the automated recognition processing results data can be sent to the keyer desk software component and the image can be permanently prevented from being displayed on the computer display. Otherwise, the image can be displayed to the human user for manual processing.

Abstract: An apparatus for integrating sensors with a traffic signal. A camera is operably disposed within a housing. The housing is attached to an object such that the camera can observe traffic flowing past a traffic signal. A visor is attached to the housing such that an optical aperture of the camera is covered by the visor, wherein the visor comprises a roof having an angle that slopes, relative to the housing, towards the optical aperture, wherein the visor further comprises a floor connected to the roof, and wherein the floor extends outwardly from the housing.

Abstract: A method and system for tracking a guidewire in a fluoroscopic image sequence is disclosed. In order to track a guidewire in a fluoroscopic image sequence, guidewire segments are detected in each frame of the fluoroscopic image sequence. The guidewire in each frame of the fluoroscopic image sequence is then detected by rigidly tracking the guidewire from a previous frame of the fluoroscopic image sequence based on the detected guidewire segments in the current frame. The guidewire is then non-rigidly deformed in each frame based on the guidewire position in the previous frame.

Abstract: A method of pattern and image recognition and identification includes building a data store of known patterns or images having known attributes and comparing those patterns to unknown patterns. The data store and comparison processing may be distributed across processors. A digital pattern recognition engine on each of the processors has the ability to compare a known pattern from the data store and an unknown pattern and compare the two patterns to determine whether the patterns constitute a match based on match criteria. If the comparison indicates a match, the match may be communicated to the data store and added as a known pattern with detected attributes to the data store. If the comparison does not indicate a match, the pattern may be flagged, transmitted to manual recognition, or further processed using character thresholding or cutting or slicing the pattern.

Abstract: Systems and methods are provided for capturing and processing images of remittance coupons using a mobile device and obtaining data from the captured image which is used to set up or carry out payment of a bill that corresponds to the remittance coupon. Optimization and enhancement of image capture and image processing are provided on the mobile device to improve the initial quality of the captured image and provide a user with real time feedback. The image is then sent from the mobile device to a remote server, where additional image processing is performed to improve the quality of the image and then extract data from the image that is relevant to paying the bill. The extracted data may be verified through comparisons with databases which store information on billers, bill formats and other relevant content that will appear on the bill.

Abstract: Systems and methods are provided for capturing and processing images of remittance coupons using a mobile device and obtaining data from the captured image which is used to set up or carry out payment of a bill that corresponds to the remittance coupon. Optimization and enhancement of image capture and image processing are provided on the mobile device to improve the initial quality of the captured image and provide a user with real time feedback. The image is then sent from the mobile device to a remote server, where additional image processing is performed to improve the quality of the image and then extract data from the image that is relevant to paying the bill. The extracted data may be verified through comparisons with databases which store information on billers, bill formats and other relevant content that will appear on the bill.

Abstract: Markers are laid out to express a predetermined code string by a plurality of continuous markers. The markers which form a marker string include at least two types of markers having different features which can be identified by image processing. Markers detected in the captured image are converted into numerical information, and are collated based on the numerical information and the predetermined code string. Using information associated with the markers, the coordinate information of each collated marker is calculated. The position and orientation of an image capturing device are calculated using the coordinate information. With a simple arrangement, the position and orientation of an image capturing device can be calculated.

Abstract: Disclosed is a method 1201 of processing a video stream, the method comprising the steps of determining 1230 a representative age measure from a model for a visual element from the video stream, determining 1250 a representative activity count measure from the model, establishing a functional relationship between the representative activity count measure and the representative age measure, comparing 1240 the functional relationship to a threshold value, and determining 1260 if the visual element is stationary, based on the result of the comparing step.

Abstract: A method of processing uniform mailpieces referred to as a “run” of mailpieces, during which method OCR is performed for recognizing certain information in a zone of interest of an image of each mailpiece, and during which method the following steps are performed: a) initializing a matrix accumulator associated with said run and including unitary accumulation elements that correspond to the pixels of the image; b) consolidating said matrix accumulator by incrementing certain unitary accumulation elements by deriving an indication of the spatial position of a block of pixels in which said certain information has been recognized unambiguously, or by using construction and local graphical correlation of blocks of image pixels to derive an optical flow map indicating local graphical movements; and c) defining, in the OCR processing, said zone of interest on the basis of the unitary accumulation elements of the consolidated matrix accumulator that present extreme accumulation values.

Abstract: An apparent motion detector is provided with multiple response levels at differing degrees of sensitivity. The motion detector is based on use of a pyroelectric infrared sensor and conventional circuitry which generates an output signal the strength of which reflects transient temperature changes occurring within a field of view and distinguishable from background heat levels. The detector's response varies with the strength of the signal using a plurality of LED's which emit different colors or are driven at different intensities. The system generally will have a field of view defined by a lens system which is translucent to visible light and transparent to infrared. The lens system doubles as a back screen projection system for display of the indicator light.

Abstract: To correct an angle error, acceleration data is received corresponding to a tracked object in a reference frame of the tracked object. Positional data of the tracked object is received from a positional sensor, and positional sensor acceleration data is computed from the received positional data. The acceleration data is transformed into a positional sensor reference frame using a rotation estimate. An amount of error between the transformed acceleration data and the positional sensor acceleration data is determined. The rotation estimate is updated responsive to the determined amount of error.

Abstract: A system and process for reading forms using a curvature distortion correction scheme is disclosed. A model image of a known form is partitioned into sectors defined by at least three alignment marks, and a transfer or correction function for each sector is generated by comparing the locations of alignment marks on a captured unread image of a form to those on a model image. The transfer function is then applied to the other marks within the sector to calculate a correct location of the marks. Each sector is sized to reasonably ensure curvature distortions are handled such that all the marks on a form are properly found and interpreted. Errors may be checked and, if they persist, smaller sectors may be defined and employed, if available.

Abstract: Mail processing methods and systems. A method includes receiving a mailpiece at a local system and scanning the mailpiece by the local system to produce an image. The method can include receiving payment information from the customer by the local system for mailing the mailpiece. The method includes applying an identification code to the mailpiece by the local system, and after the mailpiece is delivered to a central processing facility, processing the mailpiece, using the image, by a central server system at the central processing facility. Related apparatuses and systems are discussed.

Abstract: A parcel dimension measurement system includes image sensors oriented to image a parcel, an imaging subsystem configured to stitch together outputs of the image sensors to produce at least one two-dimensional image comprised of a plurality of pixels, and a general dimension subsystem including general parcel dimension information. A fine dimensioning subsystem is configured to determine dimension measurements of the parcel using the at least one two-dimensional image and the general parcel dimension information.

Abstract: A method for sorting mail includes performing an automatic address recognition process on a digitized image of a mail piece and generating a plurality of conditional address recognition results and a plurality of confirmation values each associated with one of the plurality of conditional address recognition results. The method can include sending the digitized image, the plurality of conditional address recognition results and the plurality of confirmation values to a video coding system. The method can include detecting a possible error in a first video coding result, the possible error detected using error correction information including information from the automatic address recognition process. The method may include indicating the possible error to a video coding operator, and performing a video coding operation to obtain a second video coding result and arbitrating or selecting one of the results as output.

Abstract: Secure stamping of multimedia document collections is disclosed. In one embodiment, the method comprises adding an image to a collection of media objects, and producing a visual representation of the collection that includes the image, where the visual representation has an identifier to identify the collection and graphical content representing the media objects.

Abstract: A method for retrieving a desired postal address from a plurality of postal addresses. The invention can feature receiving input data having input terms for finding the desired postal address from a first database of postal addresses where each postal address is formed of a plurality of postal address elements, searching a dictionary for entries in the dictionary corresponding to the input terms, determining that each entry in the dictionary that corresponds to the input terms corresponds to postal address elements of one or more postal addresses in the first database, and outputting data representing the desired postal address. The desired postal address is found in the first database if an address in that database has postal address elements that correspond with each of the entries of the dictionary that corresponds to one or more of the input terms.

Abstract: A method of detecting bunched-together overlapping postal items includes the steps of moving the postal items on edge above a camera, forming grayscale images including the bottom sides of the postal items, and analyzing the image for the purpose of detecting the presence of any overlapping postal items. The method further includes the steps of transforming each grayscale image into a binary image that causes a boundary defined by a “pale” zone to appear that represents the bottom side of at least one postal item and that extends between two “dark” zones, the boundary having a “right” first profile and a “left” second profile, and the method further includes the step of analyzing the right and left profiles for the purpose of detecting any profile discontinuity that is indicative of the presence of overlapping postal items.

Abstract: A universal postal code system in which mail piece delivery address information is associated with a universal postal code affixed to a mail piece. As part of the association process, the delivery address information can be verified and formatted according to postal authority standards to help ensure reliable and efficient delivery. The universal postal codes are reusable after completion of delivery by reassigning a universal postal code to a new recipient. The universal postal codes permit mailing of a mail piece without having any human readable delivery or return address information visible on the mail piece. Universal postal codes can be used for military mail systems and systems requiring anonymous mailing. Universal postal codes can also be used for automatic tabulation of votes while maintaining voter anonymity.

Abstract: The inventive method is characterized in that the prepaid postage indicia are checked to see if they contain a predefined security feature. If they do not, at least one selected read area of the mail is checked to determine whether it contains a predefined graphic representation of a prepaid postage indicia.

Abstract: A method of handling mail items includes, during a first sorting pass, taking a digital picture of each mail item, directing the mail item towards a sorting outlet bin, and deriving from the picture a digital fingerprint, and, during a second sorting pass, taking a digital picture of a current mail item to derive a fingerprint, performing automatic address recognition, and searching recorded address data for a match with the address data of the current mail item to identify a current bin to which the current mail item corresponds, so that, in order to search for a match between fingerprints, the search for matching fingerprints is conducted in an exploration space formed by the digital fingerprints that correspond to the current sorting outlet bin.

Abstract: The method of processing postal items comprises the following steps: causing the postal items to advance in series past a camera (120) for the purpose of generating an image (1) of one face of each postal item, which face bears postal address information; filtering the image of a current postal item for the purpose of isolating, in the image, at least one region of interest (ROI) containing the address information; and sending the filtered image to an automatic recognition unit (130) for automatically recognizing postal addresses by means of OCR so as to extract the postal address from said filtered image and so as to direct the current postal item to a sorting outlet. In order to construct a filter mask, the method further comprises causing a card (20) of the “separator” type to advance past the camera (120), which card has a face on which said region of interest (ROI) is represented.

Abstract: The invention relates to a method for monitoring postage indicia on mail items. According to the invention, graphic information is captured in at least one selected read area of the mail, whereupon the stored graphic information is compared to the graphic representation of the postage indicia. If the graphic information does not match one of the stored images, said mail is subjected to another monitoring step. The invention also relates to a device which is suitable for carrying out the inventive method.

Abstract: A universal cost minimizing mail arbitrator (UCMA) selects an optimum zip code result from a plurality of zip code results received from a plurality of mail processing directory match modules (DMM). The UCMA primarily has a cost tuner and an arbitration controller. The cost tuner receives truthed input for truthed mail, and user-defined cost profile parameters. The cost tuner determines an average weighted cost (AWC) for each DMM based upon the truthed input, to form a cost profile. The AWC is sorted in the cost profile based upon the Level of Sort (LOS) for the zip code result. The cost tuner gathers the cost profiles for each DMM together to form a cost profile array for all the DMMs. The arbitration controller receives the zip code results from the DMMs for each LOS code, and determines the lowest cost zip code result from amongst the various LOS codes based upon the AWC in the cost profile array. The arbitration controller selects the optimum zip code result based upon the lowest cost zip code result.

Abstract: Improvements are disclosed for further processing of a mail piece initially rejected from an automatic sorting machine. Supplemental automated processes are applied to read a bar code from the front of the piece, or read an ID Tag from the back of the piece. New labels can be applied to the front and or back sides of the piece automatically. Newly resolved address information can be applied the piece, following which it preferably is returned into the main sorting stream so that manual sorting is reduced, resulting in cost savings.

Abstract: A computer implemented system and method for payment remittance receives a payor's payment information, processes this payment information, other than a received zip code, to identify an eleven digit zip code for a payee. The system and method then accesses a database of payee records to locate a payee record corresponding to the eleven digit zip code.

Abstract: A method for providing a common framework for area of interest (AOI) specification and access. The method includes providing a first set of instructions which generates an area of interest (AOI) defined by a first geometric shape and defining the first geometric shape by one or more coordinates. The one or more coordinates are then converted to a second set of coordinates for use with a second set of instructions. The second set of coordinates is defined by a new AOI which includes information associated with the first set of instructions and which is interpreted by the second set of instructions.

Abstract: Apparatus and methods consistent with the present invention provide for identifying and processing mail using an identification code on a mailpiece as a redundant source of identification information in a mail sorting system. In one embodiment, this information is stored in a temporary database and used for the identification and processing of mail in a Remote Bar Code System (RBCS). In this embodiment, the identification code enables the automation of mail sorting and other processing tasks, reducing costs and delays in mail delivery services. In another embodiment, the identification and processing of mail occurs in an Identification Code Sorting (ICS) system. In this embodiment, a long-term database allows for mail sorting and other processing tasks on a national or global level.

Abstract: A system and method that enables early detection of hazardous materials, such as explosives and biological materials, in the early phases of mail handling or processing, is disclosed. The early detection of such hazardous material utilizing the system and method of this invention can be performed while the mail is being processed. The system includes at least two primary elements: an activation sub-system and a sense and analyze sub-system.

Abstract: A method identifies edges of an object in an image by analyzing an accumulator array produced via a Hough transformation of the image. The method includes the step of identifying a first cell in a first row of the array associated with a first edge of the object. The first is then searched for a second cell in the array associated with a second edge of the object, the second edge being parallel to the first edge. A second row shifted ninety degrees from the first row is then searched for a third cell in the array associated with a third edge of the object, the third edge being perpendicular to the first edge. The second row is then searched for a fourth cell in the array associated with a fourth edge in the image, the fourth edge being parallel to the third edge.

Abstract: The present invention encompasses a self-orthogonal character recognition engine for executing an iterative method employing a database of predetermined character strings. The method receives a digital representation of a character string. It then generates a proposed result string by applying to the captured digital image a predetermined recognition routine including one or more recognition subroutines. Each recognition subroutine employs an initial parameter setting. Next, if the proposed result string does not match any of the predetermined character strings in the database, the initial parameter setting of a recognition subroutine is changed to a next setting. The recognition process is then repeated using the next parameter setting to generate and test a next result string. The process can be repeated iteratively until a result string is verified or the process times out.

Abstract: The invention relates to a method for reading document entries and addresses. In this context, automatic reading is effected using the intermediate steps of layout analysis, classification of the segmented regions and interpretation of the results in a reader, and if necessary video coding. According to the invention, during the reading mode, when information has been clearly read automatically, the image data and the associated reading results and intermediate results, and/or, when information has been clearly ascertained using video coding, the image data and the reading results and intermediate results clearly ascertained in a second automatic reading process including the video coding results are stored for a current random sample. This current random sample is used to readapt the automatic reader.