Abstract: Systems and methods for processing documents based on a cardinal graph convolution network by generating cardinal graph representations representing words as single nodes with edges connected between neighbouring nodes in four cardinal directions. Features tensors are generated for nodes of the cardinal graph representation and the cardinal directions are encoded to generate an adjacency tensor having node neighbour indices. Entries of the adjacency tensor are transformed into a one-hot encoding of the node neighbour indices. Neighbourhood feature tensors are created over node indices and the features in each block may be scaled, convolved and reduced into new feature tensors.

Abstract: Improving the accuracy of predicted segmentation masks, including: extracting a ground-truth RGB image buffer and a binary contour image buffer from a ground-truth RGB image container for segmentation training; generating predicted segmentation masks from the ground-truth RGB image buffer; generating second binary contours from the predicted segmentation masks using a particular algorithm; computing a segmentation loss between manually-segmented masks of the ground-truth RGB image buffer and the predicted segmentation masks; computing a contour accuracy loss between contours of the binary contour image buffer and the binary contours of the predicted segmentation masks; computing a total loss as a weighted average of the segmentation loss and the contour accuracy loss; and generating improved binary contours by compensating the contours of the binary contour image buffer with the computed total loss, wherein the improved binary contours are used to improve the accuracy of the predicted segmentation masks.

Abstract: Current inspection processes employed for pipeline networks data acquisition aided with manually locating and recording defects/observations, thus leading labor intensive, prone to error and a time-consuming task thereby resulting in process inefficiencies. Embodiments of the present disclosure provide systems and methods for that leverage artificial intelligence/machine learning models and image processing techniques to automate log and data processing, reports and insights generation thereby reduce dependency on manual analysis, improve annual productivity of survey meterage and bring in process and cost efficiencies into overall asset health management for utilities, thereby enhancing accuracy in defect identification, analysis, classification thereof.

Abstract: Aspects of the present disclosure describe distributed fiber optic sensing (DFOS) systems, methods, and structures that advantageously provide traffic monitoring, and traffic management which improves the safety and efficiency of a roadway.

Abstract: A method, system and computer program product, the method comprising: obtaining a multiplicity of images taken by one or more overhead capture devices overlooking a body of water having a surface; detecting a person depicted in at least some images from the multiplicity of images; determining for images in the images, whether the image depicts the person in substantially vertical position in the at least some images said determining performed upon one or more characteristics extracted from each image; and subject to determining that the person is not in substantially vertical position in the at least some images, taking an action.

Abstract: A data analysis computer system is provided that receives a timeseries dataset and generates implied data from the dataset. The dataset is further vectorized to reduce the dimensionality of the data. Users provide input to identify windows of data that either positively or negatively correlate to instances of a given type of occurrence within the data. The user defined windows are converted to fixed sized windows and a machine learning algorithm constructs a model from the data. The model is used to predict instances of the given type of occurrence in newly received data. Validation of the predications may be performed.

Abstract: There is provided a method for detecting a lane marking using a processor including acquiring a drive image captured by an image capturing device of a vehicle which is running, detecting an edge corresponding to a lane marking from the acquired drive image and generating an edge image based on the detected edge, detecting a linear component based on the detected edge and generating a linearly processed edge image based on the detected linear component, detecting a lane marking point corresponding to the lane marking using the generated edge image and the linearly processed edge image, and detecting the lane marking based on the detected lane marking point.

Abstract: An image processing system includes a memory storing a training image set and a reference image set, and a processor including hardware. The processor is configured to: generate an augmented image set by applying data augmentation to images included in the training image set; and determine an augmentation parameter based on a similarity between an augmentation feature statistic and a reference feature statistic, the augmentation feature statistic being a statistic of a feature of a recognition target calculated based on the augmented image set, the reference feature statistic being a statistic of a feature of the recognition target calculated based on the reference image set.

Abstract: The technology disclosed performs hand pose estimation on a so-called “joint-by-joint” basis. So, when a plurality of estimates for the 28 hand joints are received from a plurality of expert networks (and from master experts in some high-confidence scenarios), the estimates are analyzed at a joint level and a final location for each joint is calculated based on the plurality of estimates for a particular joint. This is a novel solution discovered by the technology disclosed because nothing in the field of art determines hand pose estimates at such granularity and precision. Regarding granularity and precision, because hand pose estimates are computed on a joint-by-joint basis, this allows the technology disclosed to detect in real time even the minutest and most subtle hand movements, such a bend/yaw/tilt/roll of a segment of a finger or a tilt an occluded finger, as demonstrated supra in the Experimental Results section of this application.

Abstract: A method, a device and a security system for image data processing based on VR or AR are disclosed. In one aspect, an example image data processing method includes reconstructing, based on three-dimensional (3D) scanned data of a containing apparatus in which objects are contained, a 3D image of the containing apparatus and the objects contained in the containing apparatus. The reconstructed 3D image is stereoscopically displayed. Manipulation information is determined for at least one of the objects in the containing apparatus based on positioning information and action information of a user. At least a 3D image of the at least one object is reconstructed based on the determined manipulation information. The at least one reconstructed object is presented on the displayed 3D image.

Abstract: A system for displaying medical imaging data comprising one or more data inputs, one or more processors, and one or more displays, wherein the one or more data inputs are configured for receiving first image data generated by a first medical imaging device, wherein the first image data comprises a field of view (FOV) portion and a non-FOV portion, and the one or more processors are configured for identifying the non-FOV portion of the first image data and generating cropped first image data by removing at least a portion of the non-FOV portion of the first image data, and transmitting the cropped first image data for display in a first portion of the display and additional information for display in a second portion of the display.

Abstract: In one embodiment, an automated high-speed X-ray inspection system may generate a first X-ray image of an inspected sample at a first direction substantially orthogonal to a plane of the inspected sample. The first X-ray image may be a high-resolution grayscale image. The system may identify one or more elements of interest of the inspected sample based on the first X-ray image. The first X-ray image may include interfering elements that interfere with the one or more elements of interest in the first X-ray image. The system may determine one or more first features associated with respective elements of interest based on variations of grayscale values in the first X-ray images. The system may determine whether one or more defects are associated with the respective elements of interest based on the one or more first features associated with the element of interest.

Abstract: A signal processing device includes a first frame buffer configured to store a first frame, a second frame buffer configured to store a second frame and a processor. The processor is coupled to the first frame buffer and the second frame buffer and is configured to perform a first image processing procedure according to the first frame and the second frame to obtain a super resolution difference value corresponding to each pixel of the first frame, perform a second image processing procedure according to the first frame and the second frame to obtain a noise reduction value corresponding to each pixel of the first frame, selectively add the super resolution difference value and the noise reduction value to the corresponding pixel of the first frame to generate an output frame and store the output frame in the second frame buffer as the second frame.

Abstract: A chip recognition system in which a chip is configured to at least partially have a specific color indicative of a value of the chip includes: a recording device that uses a camera and records a state of the chip as an image; an image analysis device that subjects the image so recorded to image analysis and recognizes at least the specific color and a reference color that is present in the image and differs from the specific color; and a recognition device at least including an artificial intelligence device that uses a result of the image analysis by the image analysis device and specifies the specific color of the chip, wherein the artificial intelligence device of the recognition device has been subjected to teaching using, as training data, a plurality of images of the chip and the reference color irradiated with different illumination intensities.

Abstract: For an ultrasound image acquired by an ultrasound detection system, background data with small noises are first filtered. Next, a binary image is generated by performing image binarization on the noise-reduced ultrasound image based on a first threshold value, wherein the binary image contains information associated with the body of high-echo foreground images in the ultrasound image. An output image is generated by performing foreground expansion on the binary image based on the pixel value of the ultrasound image and a second threshold value smaller than the first threshold value, wherein the output image contains information associated with not only the body but also the outline of high-echo foreground images in the ultrasound image. An improved ultrasound image is generated by performing a post-processing on the ultrasound image according to information of foreground and non-foreground region in the output image.

Abstract: The present invention relates to the technical field of detection of the appearance of electronic components, and in particular to a method for detecting the appearance of six sides of a chip multi-layer ceramic capacitor based on artificial intelligence. In the method for detecting the appearance of six sides of a chip multi-layer ceramic capacitor based on artificial intelligence as provided by the present invention, a picture is automatically divided, by AI, into regions which are then classified, and extracted features are judged by conventional algorithms. It is more flexible to deal with various defects. Defect missing is avoided, and the false rate is reduced.

Abstract: An apparatus for visualizing physical movements includes: a device for acquiring video image files; a data analysis system including processor and memory; a computer program operating in the processor to identify an area in the images where periodic motions associated with physical movement of an object may be detected and quantified, and compute a new image sequence in which the motions are visually amplified; and, a user interface that displays the motion-amplified video image of the mechanical component. An associated method for using the apparatus is also disclosed.

Abstract: A chip recognition system in which a chip is configured to at least partially have a specific color indicative of a value of the chip includes: a recording device that uses a camera and records a state of the chip as an image; an image analysis device that subjects the image so recorded to image analysis and recognizes at least the specific color and a reference color that is present in the image and differs from the specific color; and a recognition device at least including an artificial intelligence device that uses a result of the image analysis by the image analysis device and specifies the specific color of the chip, wherein the artificial intelligence device of the recognition device has been subjected to teaching using, as training data, a plurality of images of the chip and the reference color irradiated with different illumination intensities.

Abstract: Example embodiments relate to hierarchical rule clustering. The examples disclosed herein access information about a set of rules, where information for an individual rule comprises information about a set of hypershapes associated with the individual rule. A respective hypervolume for each set of hypershapes associated with each individual rule may be calculated based on the accessed information. A first rule and a second rule may be combined as a new individual rule in the set of rules based on overlaps between the calculated hypervolumes.

Abstract: An imaging processing system includes an image pickup device that has circuitry which creates, from raw image data, image data including a first gamut, and compression-codes the image data including the first gamut to generate compression-coded image data. The system also includes an image processing device that has circuitry which decodes the compression-coded image data to generate uncompressed image data, which includes the first gamut. The circuitry of the image processing device also converts the uncompressed image data including the first gamut into image data including a second gamut, where the first gamut encompasses the second gamut.