Abstract: An image processing apparatus includes a processor configured to extract a component related to luminance of each of a sample image and a processing target image that is to undergo image processing to match an impression of the processing target image to the sample image, extract feature values of the processing target image and the sample image by attaching to a pixel value of each pixel forming the processing target image and the sample image a weight responsive to the component related to the luminance, and make adjustment to match the feature value of the processing target image to the feature value of the sample image.

Abstract: Methods and systems for detecting and predicting anomalies include processing frames of a video stream to determine values of a feature corresponding to each frame. A feature time series is generated that corresponds to values of the identified feature over time. A matrix profile is generated that identifies similarities of sub-sequences of the time series to other sub-sequences of the feature time series. An anomaly is detected by determining that a value of the matrix profile exceeds a threshold value. An automatic action is performed responsive to the detected anomaly.

Abstract: Aspects of the disclosure generally relate to connecting panoramic images. One or more computing devices may load and display a first panoramic image captured at a first location and receive a selection of an area on the first panoramic image, the area corresponding to where a connection to other panoramic images may be made. The one or more computing devices may identify and display one or more nearby panoramic images which were captured near the first location and receive a selection of one of the one or more nearby panoramic images. The one or more computing devices may display the selected panoramic image and the first panoramic image and align the selected panoramic image with the first panoramic image such that the selected panoramic image is oriented in the same direction as the first panoramic image. The one or more computing devices may connect the selected panoramic image with the first panoramic image.

Abstract: A processor unit configured to identify blocks of a frame of a video sequence to be excluded from a motion-compensated operation, the processor unit comprising: a frame processor configured to process pixel values of a first frame to characterise blocks of one or more pixels of the first frame as representing at least a portion of a graphic object; a frame-difference processor configured to determine difference values between blocks of the first frame and corresponding blocks of a second frame, and to process said difference values to characterise blocks of the first frame as representing an image component that is static between the first and second frames; a block identifier configured to identify blocks of the first frame as protected blocks in dependence on blocks characterised as: (i) representing at least a portion of a graphic object; and (ii) representing an image component that is static between the first and second frames, wherein the identified protected blocks are to be excluded from the motion co

Abstract: An image processing method is provided. The method includes acquiring a video. The method includes using an object detection engine to detect a person in the video. The object detection engine is integrated with an image signal processing pipeline. The method includes transmitting the video over a network. The method includes determining that the detected person has moved less than a pre-set distance. The method includes, responsive to the determining, pausing transmission of the video. An embedded image processor including an object detection engine is also provided.

Abstract: Aspects of the disclosure provide a method for denoising an image. The method can include receiving an acquired image from an image acquisition system, and processing the acquired image with a nonlinear diffusion coefficient based filter having a diffusion coefficient that is calculated using gradient vector orientation information in the acquired image.

Abstract: System, device, and method for generating and utilizing content-aware metadata, particularly for playback of video and other content items. A method includes: receiving a video file, and receiving content-aware metadata about visual objects that are depicted in said video file; and dynamically adjusting or modifying playback of that video file, on a video playback device, based on the content-aware metadata. The modifications include content-aware cropping, summarizing, watermarking, overlaying of other content elements, modifying playback speed, adding user-selectable indicators or areas around or near visual objects to cause a pre-defined action upon user selection, or other adjustments or modification. Optionally, a modified and content-aware version of the video file is automatically generated or stored. Optionally, the content-aware metadata is stored internally or integrally within the video file, in its header or as a private channel; or is stored in an accompanying file.

Abstract: Methods and systems are provided for controlling an x-ray imaging system. In one embodiment, a method for an x-ray imaging system includes acquiring, with the x-ray imaging system, a first image as an x-ray tube current of the x-ray imaging system is ramping to a target x-ray tube current, determining a corrected brightness of the first image, the corrected brightness including a measured brightness of the first image corrected by a feedback x-ray tube current relative to the target x-ray tube current, and updating the target x-ray tube current based on the corrected brightness of the first image.

Abstract: Devices, systems, and methods of generating and providing a target topographic map for finishing a photomask blank are disclosed. A method includes receiving topographic data corresponding to an uncompleted photomask blank, receiving functional specifications for flatness of an acceptable photomask blank, and generating the target topographic map for first and/or second major surfaces of the blank, which provides instructions for removing material from the first and/or second major surfaces such that the first and second major surfaces achieve a flatness that passes each functional specification. The amount of material removed reflects a reduction in material necessary to pass the functional specifications.

Abstract: Methods and systems are disclosed using camera devices for deep channel and Convolutional Neural Network (CNN) images and formats. In one example, image values are captured by a color sensor array in an image capturing device or camera. The image values provide color channel data. The captured image values by the color sensor array are input to a CNN having at least one CNN layer. The CNN provides CNN channel data for each layer. The color channel data and CNN channel data is to form a deep channel image that stored in a memory. In another example, image values are captured by sensor array. The captured image values by the sensor array are input a CNN having a first CNN layer. An output is generated at the first CNN layer using the captured image values by the color sensor array. The output of the first CNN layer is stored as a feature map of the captured image.

Abstract: A document scanner with the capability of landing a document by determining its location on the scanning bed, using intelligent methodologies for framing the document to determine the edges/boundaries of the document, processing the image using various techniques and methodologies to speed up the processing, so that the document feed track does not have to be slowed, and a unique binarization of the image to efficiently create different renderings of the scanned image.

Abstract: A video processing apparatus includes a registration component that registers configuration information associated with the video information; a filter component that filters the video frame such that all but the area of interest is excluded in a filtered video frame; a configuration component that configures a plurality of neural networks in at least one of a parallel configuration, sequential configuration, mixed parallel and sequential configuration that provides a configured plurality of neural networks; a processing component that processes the filtered video frame using the configured plurality of neural networks that provides insight information; a display that provides insight information to a user; and a storage component that stores the configuration information and insight information in persistent cloud-based storage. A corresponding method of processing video information and computer-readable medium are also disclosed.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for semi-supervised training of an object recognition model. The methods, systems, and apparatus include a monitoring system including a camera located at a property and configured to generate images and one or more computers and one or more storage devices storing instructions that are operable, when executed by the one or more computers, to cause the one or more computers to perform actions of determining a cluster of images meets a threshold for number of included images and a threshold for cluster tightness. A representative image of the cluster is selected and a query including the representative image of the cluster is provided. User feedback responsive to the query is received and an object recognition model is updated based on the user feedback.

Abstract: A method includes executing an encoder machine learning model on multiple token values contained in a document to create an encoder hidden state vector. A decoder machine learning model executing on the encoder hidden state vector generates raw text comprising an entity value and an entity label for each of multiple entities. The method further includes generating a structural representation of the entities directly from the raw text and outputting the structural representation of the entities of the document.

Abstract: Techniques are described for a multiple-phase process that uses machine learning (ML) models to produce a texturized version of an input image. During a first phase, using a pix2pix-based ML model, an automatically-generated image that depicts structured texture is generated based on an input image that visually identifies a plurality of image areas for the structured texture. During a second phase, a neural style transfer-based ML model is used to apply the style of a style image (e.g., a target image from the training dataset of the pix2pix-based ML model) to the texture image generated at the first phase (the content image) to produce a modified texture image. According to an embodiment, during a third phase, the generated texture image produced at the first phase and the modified texture image produced at the second phase are combined to produce a structured texture image with a moderated amount of detail.

Abstract: An image processing system comprises a template matching engine (TME). The TME reads an image from the memory; and as each pixel of the image is being read, calculates a respective feature value of a plurality of feature maps as a function of the pixel value. A pre-filter is responsive to a current pixel location comprising a node within a limited detector cascade to be applied to a window within the image to: compare a feature value from a selected one of the plurality of feature maps corresponding to the pixel location to a threshold value; and responsive to pixels for all nodes within a limited detector cascade to be applied to the window having been read, determine a score for the window. A classifier, responsive to the pre-filter indicating that a score for a window is below a window threshold, does not apply a longer detector cascade to the window before indicating that the window does not comprise an object to be detected.

Abstract: A method for determining an ROI in medical imaging may include receiving first position information related to a body contour of a subject with respect to a support from a flexible device configured with a plurality of position sensors. The flexible device may be configured to conform to the body contour of the subject, and the support may be configured to support the subject. The method may also include generating a 3D model of the subject based on the first position information. The method may further include determining an ROI of the subject based on the 3D model of the subject.

Abstract: An image captured by a camera can be processed by a scanning application to identify multiple regions within the image that are suitable for scanning. These regions can be detected and selected for scanning automatically. The captured regions for the single image can be stored as individual image content files.

Abstract: Systems and methods of automating the generation of an estimate of the first floor elevation of a building, using a Digital Terrain Model (DTM) and an image of the front of the building. First, machine learning (ML) algorithms are used to estimate the height of the first floor of the building above ground from the image. The estimated height is added to a DTM height of the geographical area corresponding to the location of the building to provide an estimate of the FFE. The disclosed embodiments can be applied in estimating the first floor height of any kind of building. For example, single-family homes, townhomes, high rise buildings, stores, malls, shops, movie theaters, or any other building. Additionally, the disclosed embodiments are camera-agnostic and can be applied on any type of image collected by any type of camera.

Abstract: The present disclosure provides an optical flow calculation method for a computing device, including: acquiring an event data flow with a predetermined duration from a DVS, the event data flow including a coordinate position and a timestamp of a triggered event; generating a timestamp matrix in accordance with the coordinate position and the timestamp of the triggered event; scanning elements in the timestamp matrix in a predetermined scanning direction, so as to determine at least one intermediate point in each element in accordance with a value and a gradient of the element in the predetermined scanning direction; and calculating a distance between adjacent intermediate points and a gradient direction, and generating an optical flow matrix in accordance with a calculation result. The present disclosure further provides the computing device.