Abstract: A gas detector for revealing a target gas includes an image capturing unit with multiple optical channels, an image processing unit and a calculation unit. The image processing unit is adapted for deducing a value of a radiation transmission coefficient which relates to an analysis spectral band, and which is attributable to a quantity of the target gas present in a part of the field-of-view. Preferably multiple analysis bands are used in parallel. The calculation unit is adapted for deducing an evaluation of the quantity of the target gas based on the value of the radiation transmission coefficient which relates to each analysis band. Such a gas detector may have small dimensions, be easily transportable, including on board a drone, and can provide evaluation results for the quantity of the target gas in real-time or nearly real-time.

Abstract: A method and system for segmenting a plurality of images. The method comprises the steps of segmenting the image through a novel clustering technique that is, generating a composite depth map including temporally stable segments of the image as well as segments in subsequent images that have changed. These changes may be determined by determining one or more differences between the temporally stable depth map and segments included in one or more subsequent frames. Thereafter, the portions of the one or more subsequent frames that include segments including changes from their corresponding segments in the temporally stable depth map are processed and are combined with the segments from the temporally stable depth map to compute their associated disparities in one or more subsequent frames. The images may include a pair of stereo images acquired through a stereo camera system at a substantially similar time.

Abstract: A method for enhancing detection of a fraudulent identity document in an image is provided that includes receiving, by an electronic device, an image of an identity document associated with a user including at least one background object. The method also includes determining the size and orientation of the at least one background object based on the received image, extracting information about the received image from the received image, and extracting information from the received image about a capture device that captured the received image. Each of the size and orientation of the at least one background object, the extracted received image data information, and the extracted capture device information is compared against corresponding information in record data of the user. A similarity score is calculated for each comparison. When each similarity score satisfies a threshold value, the identity document in the received image is deemed to be fraudulent.

Abstract: In some implementations a neural network is trained to perform a main task using a clustering constraint, for example, using both a main task training loss and a clustering training loss. Training inputs are inputted into a main task neural network to produce output labels predicting locations of the parts of the objects in the training inputs. Data from pooled layers of the main task neural network is inputted into a clustering neural network. The main task neural network and the clustering neural network are trained based on a main task loss from the main task neural network and a clustering loss from the clustering neural network. The main task loss is determined by comparing differences between the output labels and the training labels. The clustering loss encourages the clustering network to learn to label the parts of the objects individually, e.g., to learn groups corresponding to the object parts.

Abstract: A system including a computing device and camera is disclosed; the system configured for measuring three-dimensional attributes and associated performance measurements of a mechanical device. Some embodiments comprise a camera configured to capture images of the mechanical device and a computing device in communication with the camera. In some embodiments, the computing device is configured to access a first set of pixels associated with a first plurality of fiducials to calibrate a spatial resolution of the camera. A second image from the camera can be converted into a second set of pixels associated with each of the plurality of fiducials, which are attached to the mechanical device. The computing device can be further configured to compare the first and second set of pixels to determine the location of the plurality of fiducials on the mechanical device.

Abstract: System and methods and computer program code are provided to perform a commissioning process comprising capturing, using an image capture device, an image of an area containing at least a first fixture, identifying location and positioning information associated with the image, performing image processing of the image to identify a location of the at least first fixture in the image, and converting the location of the at least first fixture in the image into physical coordinate information associated with the at least first fixture.

Abstract: The present invention discloses a method for point cloud up-sampling based on deep learning, including: obtaining training data including a first number of sparse input points and a second number of dense input points; constructing a deep network model to be used for respectively performing replication and sampling operation based on curvature on initial eigenvectors extracted from the first number of sparse input points to obtain a second number of intermediate eigenvectors, performing splicing operation on each intermediate eigenvector, inputting the spliced intermediate eigenvectors into a multilayer perceptron, and determining sampling prediction points based on the sampling eigenvectors output by the multilayer perceptron; training the deep network model until an objective function determined by the sampling prediction points and the dense input points converges; and testing the deep network model to obtain point cloud data of an object under test after up-sampling.

Abstract: A method and apparatus for measuring a distance to an object, be a device (100) having at least two cameras (104, 106), is described. One or more first images including the object are acquired by a first camera of the device and one or more first reference images are acquired by a second camera of the device, while the device is in a first position. One or more second images including the object and one or more second reference images are acquired by cameras of the device, while the device is in a second position, different from the first position. Based on the first and second reference images, information on the displacement of at least one camera of the device between the first and second position are determined. The distance from the device to the object is calculated based on the first and second images including e object and the determined information on the displacement of the at least one camera.

Abstract: An apparatus for acquiring surrounding information of a vehicle includes: a camera configured to acquire an entire image of at least one surrounding vehicle; and a controller configured to derive at least one of coordinates of a wheel image area or coordinates of a front-rear image area included in an entire image area, and determine distance information from the vehicle to the at least one surrounding vehicle based on a relative positional relationship between the entire image area and the at least one of the wheel image area coordinates or the front-rear image area coordinates.

Abstract: Method for displaying super-resolution video of at least one-moving object without image artifacts, including the procedures of acquiring microscanned images of at least one moving object, a first and second subset of the images respectively forming a first and second data set, for each data set, analyzing at least a portion of the sub-set of the images for spatial and temporal information, determining a respective movement indication of the moving object according to the spatial and temporal information, in parallel to the procedure of analyzing, forming a respective super-resolution image from each data set and designating a respective bounded area surrounding the moving object, and repeatedly displaying each super-resolution image outside the bounded area a plurality of times at a video frame rate and displaying during those times within the respective bounded area, a plurality of consecutive microscanned images of the moving object at the video frame rate.

Abstract: The present disclosure relates to systems and methods for processing real-time video and detecting objects in the video. In one implementation, a system is provided that includes an input port for receiving real-time video obtained from a medical image device, a first bus for transferring the received real-time video, and at least one processor configured to receive the real-time video from the first bus, perform object detection by applying a trained neural network on frames of the received real-time video, and overlay a border indicating a location of at least one detected object in the frames. The system also includes a second bus for receiving the video with the overlaid border, an output port for outputting the video with the overlaid border from the second bus to an external display, and a third bus for directly transmitting the received real-time video to the output port.

Abstract: An assignment control apparatus includes an assignment device connected to multiple cameras and multiple specialized devices connected to the assignment device and each corresponding to one of preset multiple categories. The assignment device includes multiple determination units connected to the cameras one-to-one and each configured to determine the category of image data received from the camera to which the determination unit is connected, the category being one of the categories, and to output the image data along with category data indicating the determined category. The assignment device also includes an assignment unit configured to output the image data outputted from each specialized device to the specialized device corresponding to the category indicated by the category data outputted along with the image data.

Abstract: An information processing apparatus that inspects a state of a target object using distance information obtained by measuring the target object includes an acquisition unit configured to acquire position information indicating a position of a part of the target object based on the distance information, and an output unit configured to output, in a case where a position of a part of the target object does not satisfy a predetermined condition, based on the position information and the condition, information regarding a direction in which the target object is to be moved from a current state of the target object to a state satisfying the condition.

Abstract: A mechanism is provided for implementing an optical character recognition (OCR) error correction mechanism for correcting OCR errors. Responsive to receiving a document in which OCR has been performed, the mechanism assesses the document to identify a set of OCR errors generated by an OCR engine that performed the OCR using a set of visual embeddings. Responsive to identifying the set of OCR errors, the mechanism analyzes each character of a plurality of sentences within the document to generate a high-dimensional embedding for the characters of the plurality of sentences within the document. The mechanism then linguistically corrects each OCR error in the set of OCR error. The mechanism utilizes ground truth information and the set of visual embeddings to verify that character stream is linguistically correct. Responsive to verifying that the character stream is linguistically correct, the mechanism outputs an OCR error corrected document to a user.

Abstract: Disclosed herein are methods and systems for visually identifying anomaly events, comprising an edge node configured for applying a limited resources classifier to a plurality of images captured by imaging sensor(s) deployed to monitor a certain scene relating to a certain area to classify object(s) detected in the images, applying a trained context based Machine Learning (ML) model to classification data generated by the limited resources classifier to compute an anomaly score for potential anomaly event(s) relating to the detected object(s) based on one or more contextual attributes associated with the certain scene and transmitting one or more of the images to a remote server in case the anomaly score exceeds a threshold. The remote server is configured to further apply high performance visual analysis tool(s) to visually analyze the received image(s) in order to identify the one or more potential anomaly events.

Abstract: There is provided a method and a system configured to compensate for image distortions.

Abstract: Computer implemented methods and devices for determining dimensions or distances of head features are provided. The method includes identifying a plurality of features in an image of a head of a person. A real dimension of at least one target feature of the plurality of features or a real distance between at least one target feature of the plurality features and a camera device used for capturing the image is estimated based on probability distributions for real dimensions of at least one feature of the plurality of features and a pixel dimension of the at least one feature of the plurality of features.

Abstract: A medical image processing apparatus comprises processing circuitry configured to select a reference image from blood vessel images based on contrast images acquired in time series and conduct a transformation process on other blood vessel images such that blood vessel shapes in the other blood vessel images match a blood vessel shape in the reference image, generate a color image where a color that corresponds to a temporal change in a pixel value is assigned to each pixel based on the blood vessel images after the transformation process, and display the color image on a display.

Abstract: First scanned images of the first container are received from a scanning device that show the contents of the interior of the first container before the first container is cut and opened. Second scanned images that are of the contents of the first container after the first container is cut and opened are also received. The images are analyzed and, based upon the analysis, selective modifications to the operating parameters of the container opening machine are determined and made.

Abstract: Systems and methods for a machine learning system to learn a new skill without catastrophically forgetting an existing skill and to continually learn in a self-supervised manner during operation, without human intervention.