Abstract: An image tracking apparatus, which tracks a subject in a moving image, comprising: an image acquisition device that acquires image data indicating the moving image; a first detector that detects a first area of a subject in acquired image data to track the first area in the moving image; a second detector that detects a second area of a subject in the image data to track the second area in the moving image; and a controller that tracks the subject by switching between a tracking result of the first detector and a tracking result of the second detector. The first and the second detector operate independently. The controller tracks the subject using the tracking result of the first detector when the first area is tracked, and tracks the subject using the tracking result of the second detector when the first area is not tracked and the second area is tracked.

Abstract: The present disclosure relates to a dimension measuring device that shortens a time required for dimension measurement and eliminates errors caused by an operator.

Abstract: A computer-implemented method for determining field boundaries and crop forecasts in each field is provided. The method includes deriving vegetation indices for each geo-spatial pixel of each image of multi-spectral imagery at a plurality of points in time, constructing minimum bounding boxes for each image according to the vegetation indices, and generating, based on a neural network analysis of each image and the minimum bounding boxes, a geo-spatial plot of crops including a predicted plot of future crop usage for an area including each field in the multi-spectral imagery.

Abstract: In medical examination of breast cancer, a lesion computer-aided detection is performed in real time and with high accuracy, and a burden on a medical worker is reduced. A medical image processing apparatus that processes a medical image includes: a detection unit configured to detect a lesion candidate region; a validity evaluation unit configured to evaluate validity of the lesion candidate region by using a normal tissue region corresponding to the detected lesion candidate region; and a display unit configured to determine display content to a user by using an evaluation result.

Abstract: An image processing apparatus includes a detection unit that executes detection processing for detecting a particular object in an image, a holding unit that holds object information indicating a position and a size of the particular object on the image, a determination unit that determines whether a number of times a particular object is detected in the detection processing on one or more images reaches a predetermined value, a first setting unit that, when the number of times a particular object is detected in the detection processing on the one or more images is determined to reach the value, sets estimation areas on an image based on the object information obtained by the detection processing on the one or more images, and an estimation unit that executes estimation processing for estimating a number of the particular objects in the estimation areas.

Abstract: Systems, devices, media, and methods are presented for training a predictive model to detect objects in digital images, such as detecting whether a cotter key is present or absent in a digital photograph of a rail car coupler. The training system includes curating a plurality of training datasets, each including a number of raw images, together with a number of adjusted, augmented, and duplicate images. The predictive model includes a localization algorithm and an ensemble of models for classification. The localization algorithm is a deep convolutional neural network (CNN) which identifies a region of interest. One of more of the deep CNN classification models generates a plurality of candidate regions associated with each region of interest, thereby generating a large number of additional regions useful for training. In use, the trained predictive model is part of a detection and notification system that processes new images from the field and broadcasts a notice when an anomaly is detected.

Abstract: An operating method with goods information is applicable to an electronic device. The operation method includes: obtaining image information associated with one or more goods objects on a target electronic shelf among a plurality of electronic shelves in a network; and performing first communicating with a server for controlling the electronic shelves in the network according to either or both of the image information and feature information associated with the one or more goods objects, wherein feature information is extracted from the image information and the first communicating includes wirelessly transmitting either or both of the image information and the feature information associated with the one or more goods objects to the server.

Abstract: A system, method, and computer program product for automatically identifying a casualty and matching an electronic health record (EHR) to the casualty. A casualty identification is determined by matching a presenting image of the casualty with one of a master image or a social media profile image of the casualty. A recognized patient messaging system (RPMS) is configured to query one or more electronic health records (EHR) service providers for the existence of an EHR corresponding the identified casualty and automatically communicate the existence of the EHR to the EMS provider. A blockchain trusted identification module (BTIM) is configured to establish a trust relationship between the EMS provider and the one or more EHR service providers to establish a trusted pathway for delivery of the casualty's EHR to the EMS provider. With the casualty's EHR emergency responders to can provide better care for the casualty in an emergency situation.

Abstract: A method for distinguishing between types of intraoral areas of interest (AOIs) includes receiving or generating a first 3D model comprising teeth; receiving or generating a second 3D model comprising the teeth; comparing the first 3D model to the second 3D model; determining a difference between the first 3D model and the second 3D model at a dental site comprising a tooth; determining an intraoral AOI that is associated with the difference; determining whether the difference comprises a first type of change indicative of a first type of intraoral AOI or a second type of change indicative of a second type of intraoral AOI; and responsive to determining that the difference comprises the first type of change, determining that the intraoral AOI is the first type of intraoral AOI and generating an indicator of the first type of intraoral AOI for the tooth.

Abstract: A position information estimating method and apparatus are provided, wherein the position information estimating method includes estimating initial position information of a vehicle based on motion information of the vehicle that is obtained from one or more sensors, calculating search position information of one or more objects included in an image of surroundings of the vehicle based on the initial position information, evaluating a final reliability of each of the one or more objects based on the initial position information and the search position information, obtaining a filter adaptation parameter corresponding to a weight of the search position information based on the final reliability, and estimating final position information of the vehicle based on the initial position information, the search position information, and the filter adaptation parameter.

Abstract: The disclosure discloses a power transformer winding fault positioning method based on deep convolutional neural network integrated with visual identification, including 1) a winding equivalent circuit is established, and a transfer function thereof is calculated; 2) a sine wave excitation source is set at one end of the power transformer winding to obtain the amplitude-frequency characteristic curve of each winding node; 3) circuits under various fault statuses are subjected to scanning frequency response analysis to extract amplitude-frequency characteristics; 4) a feature matrix is established based on the obtained amplitude-frequency characteristics; 5) scanning frequency response analysis is performed on the diagnosed power transformer to form a feature matrix; 6) the feature matrix is converted into an image, simulation and historical detection data are used as a training set, and a deep convolutional neural network is input for training; 7) diagnosed transformer is subjected to fault classification and

Abstract: A three-dimensional camera of an automatic milking system records image data representing an outer surface of a teat of a dairy animal in three dimensions. Based on the image data, a processing unit performs a geometric analysis and calculates at least one size measure of the teat. A user interface presents output data reflecting the size-related classification of the teat.

Abstract: Systems for and methods of monitoring and analyzing deposit in an industrial water system are provided. The methods comprise heating a substrate while the substrate contacts industrial water in the industrial water system to form deposit on the substrate. A series of digital images of the substrate while the substrate contacts the industrial water in the industrial water system is created. A region of interest in the series of digital images of the substrate is defined. A deposit feature in the region of interest in the series of digital images of the substrate is identified. The deposit feature in the region of interest in the series of digital images of the substrate is analyzed to determine a deposit trend of the substrate in the industrial water system. Generally, the systems are configured so as to be capable of carrying out one or more of the methods.

Abstract: A computer-implemented method in which one or more processing devices perform operations may include obtaining a field image of a railcar collected from a field camera system and applying a machine-learning algorithm to the field image to generate a machine-learning algorithm output. The method may also include performing a post-processing operation on the machine-learning algorithm output to generate a filtered machine-learning algorithm output. Further, the method may include detecting a defect of the railcar using the filtered machine-learning algorithm output.

Abstract: A method of analyzing a substrate contacting a fluid present in an industrial system is provided. The method comprises creating a series of digital images of the substrate while contacting the fluid present in the industrial system. A region of interest in the series of digital images of the substrate is defined. A corrosion feature in the region of interest in the series of digital images of the substrate is identified. The corrosion feature in the region of interest in the series of digital images of the substrate is analyzed to determine a corrosion trend of the industrial system. In certain embodiments of the method, the fluid is industrial water, and the industrial system is an industrial water system.

Abstract: A system and method of automatic re-localization of a handheld scanning device in a previously mapped environment is provided. The system includes a two-dimensional (2D) scanner and one or more processors operably coupled to the 2D scanner. The one or more processors are responsive to non-transitory executable instructions for performing operations that include determining a current location of the 2D scanner in the environment relative to a location in a previously generated 2D image of the environment that was generated based on an image of a non-digital floorplan. The operations also include generating a new 2D image of at least a subset of the environment. The at least a subset of the environment includes the current location of the 2D scanner and at least one other location in the environment. The operations further include overlapping portions of the previously generated 2D image and the new 2D image.

Abstract: A method includes obtaining a point cloud captured by a depth sensor, and image data captured by an image sensor, the point cloud and the image data representing a support structure bearing a set of objects; obtaining an image boundary corresponding to an object from the set of objects; determining a portion of the point cloud corresponding to the image boundary; selecting, from the determined portion, a subset of points corresponding to a forward surface of the object; and generating a three-dimensional position of the object based on the forward surface.

Abstract: The system and method provide for identification of dynamic objects in an enclosed space and the presence of a component in a primary location. The system uses an active electro-optical 3D sensor, such as a three-dimensional time of flight camera, to identify the presence or absence of a reflected pulse, to determine, for example, proper placement of a seat belt, or a change in characteristics of a reflected pulse to determine a change in location, and thus possible movement, of a living creature in a vehicle, for example.

Abstract: An editor application configured for setting up at least one evaluation data structure is described, wherein each evaluation data structure is configured for evaluating a corresponding specific motion pattern in a sequence of image data structures. Each evaluation data structure comprises a ML model artifact configured for determining, based on input data comprising key data elements provided for at least one image data structure, class labels for each image data structure, said class labels identifying at least one of: at least one motion phase of the specific motion pattern, at least one evaluation point of the specific motion pattern. A key data element indicates a respective position of a landmark in the image data structure.

Abstract: In one embodiment, a method includes detecting objects in an image. The method includes accessing a mask for each object. The method includes receiving an input in relation to the image. The input corresponds to an input region and an input type. The method includes identifying a region of the image corresponding to the input region of the input. The identified region of the image includes one or more of the masks. The method includes providing feedback regarding the one or objects in the identified region of the image based on the input type.