Abstract: A method includes: recording an image at a sensor block, the sensor block; detecting a set of objects in the image based on an object classification model; generating a feature vector for the image based on the set of objects; accessing a workstation template for the work area, the workstation template defining a set of default objects for each workstation in the work area; identifying a subset of objects comprising an intersection of the set of default objects and the set of objects; removing the subset of objects from the feature vector; accessing an occupancy template for the work area, the occupancy template defining a set of significant objects; classifying the workstation as occupied based on the feature vector and the set of significant objects; and rendering a representation of the workstation at a work area interface, the representation indicating occupancy of the workstation.

Abstract: A vehicle number identification device includes a registration-number-information acquisition part configured to acquire registration number information associated with an identification media installed in a vehicle; a number-plate-image acquisition part configured to acquire a number-plate image of the vehicle; an OCR processing part configured to acquire OCR resultant information representing a result of an optical character recognition with respect to the number-plate image; an estimation processing part configured to input the number-plate image into a machine-learning model and to acquire estimation result information output from the machine-learning model; and a matching determination part configured to determine whether at least two types of information out of the registration number information, the OCR resultant information, and the estimation result information indicate a same vehicle number and to identify the same vehicle number as a vehicle number of the vehicle.

Abstract: A driving support method including: detecting a first boundary point group including boundary points forming a lane boundary line, using a sensor mounted on a vehicle; converting a coordinate system of the first boundary point group to the same coordinate system as a coordinate system of map data; and performing matching between a first point group constituted by boundary points not close to each other in either one of a second boundary point group including boundary points forming a lane boundary line in the map data and the first boundary point group and the other of the first boundary point group and the second boundary point group.

Abstract: An object recognition device of the present disclosure divides a reflection intensity image and a background light image acquired from a light sensor into the same number of subregions. The object recognition device calculates a feature amount for partial image data of each of the subregions of the reflection intensity image and partial image data of each of the subregions of the background light image that have been divided. The object recognition device compares the feature amounts calculated for the partial image data of the reflection intensity image and the partial image data of the background light image in the same one of the subregions and selects the feature amount with which an object is more easily recognized as the selected partial image data. The object recognition device recognizes the object based on overall image data configured by the selected partial image data pieces that have been selected.

Abstract: A human body identification method, an electronic device and a storage medium, related to the technical field of artificial intelligence such as computer vision and deep learning, are provided. The method includes: inputting an image to be identified into a human body detection model, to obtain a plurality of preselected detection boxes; identifying a plurality of key points from each of the preselected detection boxes respectively according to a human body key point detection model, and obtaining a key point score of each of the key points; determining a target detection box from each of the preselected detection boxes, according to a number of the key points whose key point scores meet a key point threshold; and inputting the target detection box into a human body key point classification model, to obtain a human body identification result for the image to be identified.

Abstract: A method, a computer program product, and a computer system compare images for content consistency. The method includes receiving a first image including a first document and a second image including a second document. The method includes performing a visual classification analysis on the first image and the second image. The visual classification analysis generates an overlap of the first image with the second image. The method includes determining whether a region of the overlap is indicative of a content inconsistency. As a result of the region of the overlap being indicative of a content inconsistency, the method includes performing a character recognition analysis on a first area of the first image and a second area of the second image corresponding to the region of the overlap to verify the content inconsistency.

Abstract: This application relates to a location tracking device and method using a feature matching. The location tracking device may include a sensor, a camera, and a controller. The sensor is provided in a predetermined object and collects sensing information including at least one of speed, direction, gravity, and acceleration of the object. The camera is provided in the predetermined object and collects image information by capturing an image. The controller calculates an initial fundamental matrix (F0) by using the collected sensing information and calibration information of the camera, detects feature points of the image information, performs a feature matching in a fundamental matrix (F) by combining the initial fundamental matrix and the feature points, and tracks a location of the object by using a result of the feature matching.

Abstract: Systems and methods are disclosed for determining anatomy directly from raw medical acquisitions using a machine learning system. One method includes obtaining raw medical acquisition data from transmission and collection of energy and particles traveling through and originating from bodies of one or more individuals; obtaining a parameterized model associated with anatomy of each of the one or more individuals; determining one or more parameters for the parameterized model, wherein the parameters are associated with the raw medical acquisition data; training a machine learning system to predict one or more values for each of the determined parameters of the parametrized model, based on the raw medical acquisition data; acquiring a medical acquisition for a selected patient; and using the trained machine learning system to determine a parameter value for a patient-specific parameterized model of the patient.

Abstract: A system and method for integrating a widget in a third-party application. Initially, a widget software code related to insurance payment capture and verification may be received. Further, the widget software code may be integrated into an application code of a third-party application. The widget software code may be executed to invoke an insurance payment widget. The insurance payment widget is configured to capture insurance card images, and extract insurance information based on an analysis of the insurance card images. Furthermore, the extracted insurance information may be validated using an insurance database and plans curated by an application provider. Subsequently, an insurance eligibility status and insurance coverage benefits of the insurance subscriber may be received and used to enable an insurance payment mode on the third party application in real-time, thereby integrating the widget in the third party application.

Abstract: This disclosure is directed to, in part, a processing pipeline for detecting predefined activity using image data, identifying an item-of-interest and a location of the item-of-interest across frames of the image data, determining a trajectory of the item-of-interest, and determining an identifier of the item-of-interest and an action taken with respect to the item-of-interest. The processing pipeline may utilize one or more trained classifiers and, in some instances, additional data to identify items that are placed into or removed from a tote (e.g., basket, cart, or other receptacle) by users in material handling facilities as the users move around the material handling facilities.

Abstract: Disclosed herein are apparatuses and methods for auto-configuring a region of interest (ROI) associated with a camera. In one implementation, a method comprises receiving image frames from a camera installed in the environment, wherein the ROI is located within the view of the image frames. The method includes tracking a plurality of persons in the image frames and determining a respective trajectory of movement for each person of the plurality of persons. The method further includes comparing each of the respective trajectories to one another and identifying, based on the comparing, a common trajectory shared by more than one person of the plurality of persons, wherein the common trajectory is not fully encompassed in the ROI. The method additionally includes updating the ROI to encompass the common trajectory, and includes configuring the updated ROI to be associated with new image frames from the camera.

Abstract: A method of automatically detecting adhesive wash-off includes: capturing, by a vision system, before-images of adhesive-covered areas on an assembly bonded by an adhesive before a process; capturing, by the vision system, after-images of the adhesive-covered areas on the assembly after the process; and determining, by a data processing module, an occurrence of the adhesive wash-off based on a comparison between the before-images and the after-images. The method further includes automatically determining a root cause of the adhesive wash-off when a number of occurrences of the adhesive wash-off exceeds a threshold.

Abstract: Examples described herein include methods and computing systems which may include examples of calculating risk scores for certain natural disasters perils based on machine learning model outputs. For example, a machine learning model may weight each of the pixels of a map in accordance with the set of weights associated with a structure, to calculate a risk score for a particular natural disaster peril associated with that structure. A plurality of risk selections may be provided to a user computing device for selection by a user, with those risk selections being associated with that risk score. Advantageously, the computing system facilitates the interaction of datasets with different measurement parameters in a machine learning model. In normalizing datasets before providing the datasets to input nodes of a machine learning model, a computing system may efficiently provide hazard and vulnerability outputs of the machine learning model.

Abstract: Methods are disclosed for the generation and editing of layer delineations within three-dimensional tomography scans. Cross sections of a subject are generated and presented to an operator, who has the ability to edit layer delineations within the cross section, or determine parameters used to generate new cross sections. By guiding an operator through a set of displayed cross sections, the methods can allow for a more rapid, efficient, and error-free segmentation of the subject. The cross sections can be nonplanar in shape or planar and non-axis-aligned. The cross sections can be restricted to exclude one or more user-defined regions of the subject, or to include only one or more user-defined regions of the subject. The cross sections can be localized to a point-of-interest. Iterative implementations of the methods can be used to arrive at a segmentation deemed satisfactory by the user.

Abstract: The present invention relates to a method for registering a first anatomical structure (1) which is articulately coupled to a second anatomical structure (2), the method being constituted to be executed by a computer and comprising the steps of: acquiring second structure correlation data describing the spatial position of at least one correlation feature (3) relative to the second anatomical structure (2); acquiring coupling data describing a positional fixation (6) of the first anatomical structure (1) relative to the second anatomical structure (2), which is set by the articulated coupling (4) between the first anatomical structure (1) and the second anatomical structure (2); determining, based on the second structure correlation data and the coupling data, first structure correlation data describing the spatial position of the at least one correlation feature (3) relative to the first anatomical structure (1). The present invention further relates to a corresponding computer program and system.

Abstract: A few-shot defect detection method based on metric learning, including: (S1) performing data enhancement on a to-be-detected few-shot defect data set through a G2-Generative adversarial network (G2-GAN); (S2) extracting features of a defect data set similar to the to-be-detected few-shot defect data set based on an adaptive convolution kernel-based convolutional neural network (SKM-CNN) to generate a pre-training model; and (S3) transferring the pre-training model to a few-shot defect detection network (S2D2N) based on metric learning; and performing target feature extraction and metric learning in sequence to realize rapid identification and location of defects.

Abstract: The present invention is directed to a helmet wearing determination system including a imaging means that is installed in a predetermined position and images a two-wheel vehicle that travels on a road; and a helmet wearing determination means that processes an image imaged by the imaging means, estimates a rider head region corresponding to a head of a person who rides on the two-wheel vehicle that travels on the road, compares image characteristics of the rider head region with image characteristics according to the head at a time when a helmet is worn or/and at a time when a helmet is not worn, and determines whether or not the rider wears the helmet.

Abstract: Described are methods for identifying the in-field positions of plant features on a plant by plant basis. These positions are determined based on images captured as a vehicle (e.g., tractor, sprayer, etc.) including one or more cameras travels through the field along a row of crops. The in-field positions of the plant features are useful for a variety of purposes including, for example, generating three-dimensional data models of plants growing in the field, assessing plant growth and phenotypic features, determining what kinds of treatments to apply including both where to apply the treatments and how much, determining whether to remove weeds or other undesirable plants, and so on.

Abstract: A correcting method of a real-time image is disclosed and includes: continuously receiving real-time images from an image inputting device; analyzing an image feature in each real-time image; computing a motion vector of each real-time image based on a feature difference between any two time-adjacent real-time images; computing a moving trajectory predicted value of a latest real-time image based on accumulated motion vectors; computing a compensation value of the latest real-time image when determining that a difference between the moving trajectory predicted value and the motion vector of the latest real-time image is within a correction allowable range, and correcting the latest real-time image by the compensation value; and, resetting the image inputting device and outputting a default image when determining that the difference between the moving trajectory and the motion vector of the latest real-time image exceeds the correction allowable range.

Abstract: Convenience is improved by enabling an imaging section to capture an image of a measurement element without adjustment of a rotation angle of a measurement object performed by a user. A control unit identifies a reference rotation angle during measurement, and calculates a measurement angle to measure a measurement element based on a relative rotation angle with respect to the reference rotation angle stored in advance. The control unit controls a rotation unit such that a rotation angle of the rotation unit becomes the measurement angle. The control unit executes a process of measuring a dimension of the measurement element based on a measurement object image captured by an imaging section when the rotation angle of the rotation unit becomes the measurement angle.