Abstract: This disclosure generally relates to method and system for asset inspection using unmanned aerial vehicles. Regular inspection of assets in any environment is time consuming and involves physical risk and injury to the property. The method includes receiving a plurality of input images of an asset using unmanned aerial vehicle and generates a 3D point cloud based on overlapped images identified from the plurality of input images of the asset being inspected. Further, anomalies are detected and defects are observed in the one or more objects of interests based on the change observed in each 2D image frame from the plurality 2D image frames in a specific view with an identical objects of interest of the 2D image frame. A pyramidal view map is created for the plurality 2D image frames by mapping each pixel of the plurality 2D image frames with the plurality of input images.

Abstract: A method to determine activity in a sequence of successively acquired images of a scene, comprises: acquiring the sequence of images; for each image in the sequence of images, forming a feature block of features extracted from the image and determining image specific information including a weighting for the image; normalizing the determined weightings to form a normalized weighting for each image in the sequence of images; for each image in the sequence of images, combining the associated normalized weighting and associated feature block to form a weighted feature block; passing a combination of the weighted feature blocks through a predictive module to determine an activity in the sequence of images; and outputting a result comprising the determined activity in the sequence of images.

Abstract: Disclosed are methods, systems, and computer-readable medium for obstacle detection and database management using multi-source weightages. For instance, the method may include: responsive to a detection of an object, determining whether the object is a known object or an unknown object based on object features and known objects indicated by vehicle state information and an obstacle database; responsive to a determination that the object is an unknown object, updating the obstacle database with unknown object information and transmitting the unknown object information to an off-board service; receiving a response from the off-board service, the response including a weightage assigned to the unknown object; updating the obstacle database with the weightage assigned to the unknown object; and performing at least one action based on the weightage of the unknown object.

Abstract: Various exemplary embodiments relate to an anonymous database system. The system includes a plurality of biometric nodes in communication with one another. Each of the plurality of biometric nodes includes a biometric input that receives biometric data from a user. The system also includes at least one central database in communication with the plurality of biometric nodes; and a plurality of institution databases in communication with the plurality of biometric nodes. A first node of the plurality of biometric nodes is configured to receive a message from a second node of the plurality of biometric nodes, the message requesting authorization of data access by the second node. Various embodiments relate to a method for performing an action requiring multiple levels of authentication using an anonymous database system.

Abstract: Systems and methods for real time feedback and for updating welding instructions for a welding robot in real time is described herein. The data of a workspace that includes a part to be welded can be received via at least one sensor. This data can be transformed into a point cloud data representing a three-dimensional surface of the part. A desired state indicative of a desired position of at least a portion of the welding robot with respect to the part can be identified. An estimated state indicative of an estimated position of at least the portion of the welding robot with respect to the part can be compared to the desired state. The welding instructions can be updated based on the comparison.

Abstract: Systems and methods for processing spatial structure data are provided. The system accesses spatial structure data, which describes object structure, and which has depth information indicative of a plurality of layers for the object structure. The system further extracts, from the spatial structure data, a portion of the spatial structure data representative of one layer of the plurality of layers. The system identifies, from the portion of the spatial structure data, a plurality of vertices that describe a contour of the layer. Additionally, the system identifies convex corners of the layer based on the plurality of vertices and performs object recognition according to the convex corners.

Abstract: Systems and method are provided for managing inventory by determining grading returned products. One method includes, generating a hierarchical model based on the return product identifier, the hierarchical model comprising a plurality of data structures; and generating an agent configured to traverse the plurality of data structures of the hierarchical model. The method may further include determining a price of the product based on a product condition; modifying the database to assign the price to the return product identifier; transmitting a notification, to the user device, to resell the product at the determined price; and providing, for presentation via the user device, a user interface that includes the notification.

Abstract: A system including circuitry configured to process image data of a meal to obtain information on the contents of the meal; generate, based on the obtained information, a query with guidance to change image capture settings: and guide a user to pick up at least a part of the meal.

Abstract: The invention relates to a method and system for performing content-aware deduplication of video files and content storage cost optimization. The method includes pre-processing video files into a plurality of groups of video files based on type of genre and run-time of a video. The genre of a plurality of video files is automatically detected using a sliding-window similarity index, which is utilized to improve accuracy of genre detection. After the pre-processing step, each group of the plurality of groups of video files are simultaneously fed into a plurality of machine learning (ML) instances and models which measure a degree of similarity corresponding to each group of video files by detecting one or more conditions that exists in the video files. The one or more conditions are detected by performing deep inspection of content in the video files using hash-based active recognition of objects.

Abstract: A processing system of a mobile device acquires an image of an object of a target color, wherein the image was captured by an integrated digital camera of the mobile device, calculates a first plurality of values that describes the target color, and wherein the calculating is based on an analysis of a pixel of the image, and identifies a first candidate color from among a plurality of candidate colors, wherein each candidate color in the plurality of candidate colors is associated with a second set of values that describes the each candidate color, and wherein the second set of values describing the first candidate color matches the first set of values more closely than any second set of values associated with another candidate color of the plurality of candidate colors.

Abstract: In various example embodiments, a system and method for integration of a three-dimensional model is disclosed. In one example embodiment, a method includes receiving a plurality of images, selecting points on the images and triangulating the points to generate a plurality of depth maps, generate a three-dimensional mesh by combining the plurality of depth maps, generating a three-dimensional model of the item by projecting the plurality of images onto the mesh using the points, calibrating colors used in the model using colors diffuse properties of the colors in the images, and providing a user interface allowing a user to select one or more user points on the three-dimensional model and provide additional information associated with the selected user points.

Abstract: An image processing method includes analyzing multiple images data based on Illumination-invariant Feature Network (IF-NET) with an image processing device to generate corresponding sets of eigenvector, in which image data includes a first image data related to at least one first feature of the sets of eigenvector, and a second image data related to at least one second feature of the sets of eigenvector; choosing a corresponding first training set of tiles and second training set of tiles from the first image data and second image data with an image processing device based on IF-NET, and computing on both training set of tiles to generate a least one loss value; and adjusting IF-NET based on a least one loss value. An image processing system is also disclosed herein.

Abstract: Embodiments herein provide a method and system for real time ROI detection in thermal face images based on a heuristic approach. The ROI of the thermal images, once detected, is then further used to detect temperature of a subject corresponding to the ROI. Unlike state of the art techniques, the heuristic approach is computationally less intensive and provides fast and accurate ROI detection even in case of occluded faces in a crowd with a single thermal image having a plurality of subject being scanned. The heuristics applied does not focus on face detection but directly on point of interest detection. Once the point of interest (ROI) is detected, it may be used for plurality of applications such as subject tracking and the like, not limited to subject or object temperature sensing since the method disclosed herein is easily implementable on low power devices.

Abstract: Various embodiments of the disclosure relate to an accelerated mathematical engine. In certain embodiments, the accelerated mathematical engine is applied to image processing such that convolution of an image is accelerated by using a two-dimensional matrix processor comprising sub-circuits that include an ALU, output register and shadow register. This architecture supports a clocked, two-dimensional architecture in which image data and weights are multiplied in a synchronized manner to allow a large number of mathematical operations to be performed in parallel.

Abstract: Disclosed are systems, devices, and methods for detecting characteristics of an unhatched egg. A set of images of an unhatched egg are obtained, where each of the spectral images is obtained in a particular wavelength range. The set of images is processed to extract image features, where the image features includes an image texture feature. The extracted image features are processed to classify the unhatched egg according to at least one characteristic. The at least one characteristic may include fertility and/or gender.

Abstract: Systems and methods for identification (ID) document verification using hybrid near-field communications (NFC) authentication and optical authentication are provided. An exemplary method includes receiving, by a client device, an image of an ID document. Based on the image of the ID document, a determination is made whether the ID document includes a near-field communications (NFC) chip that stores data comprising identifying information for an owner of the identification. Based on this determination of whether the ID document includes an NFC chip, the ID document is verified by selectively using at least one of NFC chip authentication and optical authentication, to obtain a verification result.

Abstract: A computer implemented method to present digital images may include storing a digital image in a database and applying a digital image processing technique to the digital image to identify a region of interest of the digital image. The method may also include storing region data that identifies the region of interest of the digital image in the database and receiving a request for information associated with the digital image from a digital device. In response to the request, the method may include providing the digital image and the region data for transmission to the digital device, the digital device configured to adjust a cropping view of the digital image based on the region data to display the region of interest of the digital image.

Abstract: The disclosure generally relates to methods and systems for generating synthetic microstructure images of a material with desired features. Conventional techniques that make use of unsupervised deep generative models has no control on the generated microstructure images with specific, desired set of features. The present disclosure generates the synthetic microstructure images of the material with desired feature, by using a variational autoencoder defined with a style loss function. In the first step, the variational autoencoder is trained to learn latent representation of microstructure image of the material. In the second step, some of the dimensions of learned latent representation is interpreted as physically significant features. In the third and last step, the latent representation required for getting the desired features is appropriately sampled based on the interpretation to generate the synthetic microstructure images of the material with desired features.

Abstract: A multi-layer technology stack includes a sensor layer including image sensors, a device layer, and a cloud layer, with interfaces between the layers. A method to curate different custom workflows for multiple applications include the following. Requirements for custom sets of data packages for the applications is received. The custom set of data packages include sensor data packages (e.g., SceneData) and contextual metadata packages that contextualize the sensor data packages (e.g., SceneMarks). Based on the received requirements and capabilities of components in the technology stack, the custom workflow for that application is deployed. This includes a selection, configuration and linking of components from the technology stack. The custom workflow is implemented in the components of the technology stack by transmitting workflow control packages directly and/or indirectly via the interfaces to the different layers.

Abstract: An adaptive learning power modeling method includes: sampling at least one of a plurality of network components to form a power consumption evaluation network according to at least one parameter within a parameter range; evaluating a predictive power consumption of a to-be-measured circuit by the power consumption evaluation network; training and evaluating an actual power consumption and the predictive power consumption of the to-be-measured circuit by the power consumption evaluation network to obtain an evaluation result; and performing training according to the evaluation result to determine whether to change the power consumption evaluation network.