444 results about "Pixel correlation" patented technology

An illustrative system may include a stereoscopic imaging device comprising: a first image sensor having a first pixel array and configured to provide a first windowed output comprising first image information associated with pixels included within a first window of the first pixel array, the first window having a first position; and a second image sensor having a second pixel array and configured to provide a second windowed output comprising second image information associated with pixels included within a second window of the second pixel array, the second window having a second position; and an imaging control system that may be configured to: determine a working distance between a target surface within a scene and the stereoscopic imaging device; and adjust one or more of the first position of the first window within the first pixel array or the second position of the second window within the second pixel array.

Apparatus and method for object pose estimation are disclosed. The apparatus includes at least a processor and a memory communicatively connected to the at least a processor, wherein the memory contains instructions configuring the at least a processor to receive a plurality of sets of echo data, wherein the plurality of sets of echo data is configured for generation of a plurality of echo depth maps, segment the plurality of echo depth maps, determine a depth datum related to pixels of an object of interest as a function of the plurality of segmented echo depth maps, generate a three dimensional (3D) point cloud related to the object of interest as a function of the depth datum and generate a pose datum of the object of interest as a function of the 3D point cloud.

The invention provides an unmanned aerial vehicle small target detection algorithm based on improved YOLOv8, and the algorithm comprises the steps: combining the features of a related pixel point related to a target pixel point to the target pixel point, generating an enhanced feature map, adding a CBAM module of dynamic attention, dynamically generating an attention weight, generating a redistribution feature map, and inputting a detection head of the YOLOv8. And marking a corresponding bounding box, a category and a confidence label at each to-be-detected target, and the like. Through a non-local mechanism and introduction of a dynamic CBAM module, the model can automatically adjust a feature attention area according to the specific feature of each image, in the feature extraction process, through the global pixel correlation degree in a modeling feature map, compared with a local pixel correlation module, the features of a small target can be recognized by a global view angle, and the feature extraction efficiency is improved. And the model is helped to more effectively pay attention to a fine and easily-confused target area.

This disclosure provides systems, devices, apparatus, and methods, including computer programs encoded on storage media, for exploiting spatial locality to speed up first-hit ray BVH traversals. A processor may generate a bounding volume hierarchy (BVH) hit map based a set of pixels, where the BVH hit map includes entries indicative of intersection information for a set of rays associated with the set of pixels, and where the BVH hit map corresponds to a BVH. The processor may select an entry from the entries of the BVH hit map based on pixel coordinate information for the set of pixels. The processor may traverse the BVH from a node of the BVH associated with the selected entry.

A method for detecting at least one object of interest in at least one raw data x-ray image includes the steps of emitting an incident x-ray radiation beam through a scanning volume having an object therein, detecting x-ray signals transmitted through at least one of the scanning volume and the object, deriving the at least one raw data x-ray image from the detected x-ray signals, inputting the raw data x-ray image, expressed according to an attenuation scale, into a neural network, for each pixel in the raw data x-ray image, outputting from the neural network a probability value assigned to that pixel, and, classifying each pixel in the raw data x-ray image into a first classification if the probability value associated with the pixel exceeds a predetermined threshold probability value and in a second classification if the probability value associated with the pixel is below the predetermined threshold probability value.

A vehicle external environment recognition apparatus includes at least one processor, and at least one memory coupled to the at least one processor. The at least one processor is configured to operate in cooperation with at least one program stored in the at least one memory to execute processing. The processing includes generating a distance image from luminance images, specifying, by using semantic segmentation, a floating matter class in the luminance images, and invalidating parallax associated with floating pixels that are included in the distance image and belong to the floating matter class.

Recurrent blurring may be used to render frames of a virtual environment, where the radius of a filter for a pixel is based on a number of successfully accumulated frames that correspond to that pixel. To account for rejections of accumulated samples for the pixel, ray-traced samples from a lower resolution version of a ray-traced render may be used to increase the effective sample count for the pixel. Parallax may be used to control the accumulation speed along with an angle between a view vector that corresponds to the pixel. A magnitude of one or more dimensions of a filter applied to the pixel may be based on an angle of a view vector that corresponds to the pixel to cause reflections to elongate along an axis under glancing angles. The dimension(s) may be based on a direction of a reflected specular lobe associated with the pixel.

The invention relates to the technical field of image processing, and discloses an optical image noise reduction method and system of a composite imaging assembly. The method comprises the following steps: carrying out geometric correction on visible light, infrared and laser sensor data through a multi-sensor time synchronization mechanism to obtain a noise coupling matrix; generating annular and checkerboard type mask modes based on the noise coupling matrix; performing cross-band feature interaction through the conditional mask convolution block to obtain a multi-scale dense connection feature map; a double-blind-spot self-supervised learning algorithm is adopted to optimize cross-band coordination loss, and a noise reduction prediction result is obtained; and finally, edge protection processing is carried out through a block-based Wiener filter, and a final noise reduction image is obtained. The problem of noise reduction failure caused by multi-sensor noise coupling and pixel strong correlation in the composite imaging assembly is solved, and the imaging quality and target recognition precision of the composite imaging assembly in a complex environment are improved.

Image sensors with improved memory effect are disclosed herein. In one embodiment, a method for reducing image lag associated with a pixel included in a plurality of pixels is described. The pixel includes a photodiode, a first floating diffusion coupled to the photodiode through a transfer transistor, a second floating diffusion coupled to the first floating diffusion through a dual floating diffusion transistor, and a lateral overflow integration capacitor coupled between the second floating diffusion and a bias voltage source. The lateral overflow integration capacitor is further coupled to a pixel reference voltage source through a reset transistor. Operation of the pixel comprises an idle period and an integration period after the idle period. The method also includes configuring the lateral overflow integration capacitor to be either zero-biased or forward-biased during the idle period.

In accordance with some embodiments, systems, methods and media for single-photon imaging with improved energy efficiency are provided. In some embodiments, the system comprises: an image sensor comprising pixels comprising detectors in an array, configured to: generate image data representing a scene, and comprising pixel values based on a signal generated by the respective detector during a first period of time; and a processor configured to: determine, for each pixel, an inhibition value based on the pixel value; determine, based on the inhibition value, that a pixel is to be inhibited during a second period of time; inhibit the pixel such that a pixel value associated with the pixel in second image data is consistent with the pixel being inhibited during a second period of time.

A computer-implemented method for identifying one or more changes across a plurality of images, the method comprising: receiving, at a convolutional neural network, CNN, encoder, CNN input data comprising data associated with each pixel of each of the plurality of images; propagating the CNN input data through the CNN encoder to generate a plurality of feature maps, wherein each feature map comprises a feature classification of each pixel of a respective image of the plurality of images according to a feature classification scheme, wherein the feature classification scheme comprises a plurality of classifications and is generated by the CNN encoder based on training data; receiving, at a ConvLSTM network, ConvLSTM input data comprising the plurality of feature maps generated by the CNN encoder; and propagating the ConvLSTM input data through the ConvLSTM network to generate a change map, wherein the change map comprises change data indicative of one or more changes across the plurality of images.

A farming machine includes one or more image sensors for capturing an image as the farming machine moves through the field. A control system accesses an image captured by the one or more sensors and identifies a distance value associated with each pixel of the image. The distance value corresponds to a distance between a point and an object that the pixel represents. The control system classifies pixels in the image as crop, plant, ground, etc. based on depth information in in the pixels. The control system generates a labelled point cloud using the labels and depth information, and identifies features about the crops, plants, ground, etc. in the point cloud. The control system generates treatment actions based on any of the depth information, visual information, point cloud, and feature values. The control system actuates a treatment mechanism based on the classified pixels.

Techniques disclosed herein relate to identifying individuals in digital images. In some embodiments, a digital image may be acquired (802) that captures an environment containing at least a first subject. A first portion of the digital image depicting the first subject may be segmented (806) into a plurality of superpixels. For each superpixel of the plurality of superpixels: a semantic label may be assigned (810) to the superpixel; features of the superpixel may be extracted (812); and a measure of similarity between the features extracted from the superpixel and features extracted from a reference superpixel identified in a reference digital image may be determined (814), wherein the reference superpixel has a reference semantic label that matches the semantic label assigned to the superpixel. Based on the measures of similarity associated with the plurality of superpixels, it may be determined (818) that the first subject is depicted in the reference image.

The invention provides a super-resolution binocular image generation method and system based on geometric structure consistency, and the method comprises the steps: extracting the deep features of a low-resolution binocular image through employing a convolutional neural network or a Transform model, achieving the information interaction of a left image and a right image in combination with a cross attention module, and constructing a pixel incidence matrix of the left image and the right image; acquiring a pixel corresponding relation of the left image and the right image by using the pixel incidence matrix, and constructing a continuous parallax field; performing spatial warping on the deep features based on a continuous parallax field to obtain warping features, and aligning the deep features of the left and right images; and merging the deep features and the warping features after spatial alignment, inputting the merged features into a feature up-sampling module based on implicit two-dimensional expression, and outputting a high-resolution binocular image of the same scene. The invention provides a binocular image super-resolution technology comprising binocular image feature extraction, continuous parallax field construction based on implicit two-dimensional expression, left and right image feature space alignment and feature upsampling based on implicit two-dimensional expression.

Systems and methods are provided for vehicle navigation. In one implementation, at least one processing device may receive, from a camera of the host vehicle, at least one captured image representative of an environment of the host vehicle. The processing device may analyze one or more pixels of the at least one captured image to determine whether the one or more pixels represent at least a portion of a target vehicle. For pixels determined to represent at least a portion of the target vehicle, the processing device may determine one or more estimated distance values from the one or more pixels to at least one edge of a face of the target vehicle; and generate, based on the analysis of the one or more pixels, including the determined one or more distance values associated with the one or more pixels, at least a portion of a boundary relative to the target vehicle.

A method of training a machine learning model to identify image features, the method comprising: a. providing training image data, the training image data comprising a plurality of pixels; b. assigning a groundtruth annotation to each pixel, wherein each groundtruth annotation relates to a respective one of the pixels and each groundtruth annotation indicates whether or not that the pixel corresponds with an image feature; c. providing an ignore mask comprising a set of ignore flags, wherein each ignore flag relates to a respective one of the pixels and each ignore flag provides an indication that the pixel should be ignored; d. for each pixel, receiving a prediction value from the machine learning model, wherein each prediction value provides an indication of a probability of the pixel corresponding with an image feature; e. for each pixel which has no ignore flag, determining a loss value based on the prediction value and groundtruth annotation for that pixel, and f. training the machine learning model on a basis of the loss value; and for each pixel which has an ignore flag, ignoring the prediction value for that pixel so that it is not used to train the machine learning model.

An event filter for an event based imaging sensor is provided by the present disclosure, the event filter comprising: input circuitry configured to receive an indication of a pixel position associated with a current event detected by the event based imaging sensor; event potential determining circuitry configured to determine an event potential associated with the current event; and filtering circuitry configured to determine whether to pass the current event based on the event potential; wherein the event potential determining circuitry is configured to determine the event potential for the current event based on combining contributions associated with pixels in a local region of pixels around the pixel position of the current event, where a given contribution associated with a given pixel in the local region is dependent on a time decay factor which causes the given contribution to decrease with increasing time difference between a time of detection of a previous event for the given pixel and a time of detection of the current event. A method of filtering and a computer program are also provided.

A macropixel is disclosed. The macropixel comprises a plurality of pixels, each pixel comprising a Single Photon Avalanche Diode (SPAD). The macropixel also comprises a memory configured to store: a plurality of counts, each count associated with a pixel of the plurality of pixels; and a plurality of saturation bits, each saturation bit associated with a count of the plurality of counts. The macropixel also comprises saturation detection circuitry configured to gate a recharge of each SPAD based on a state of the respective saturation bit.

The invention discloses a video image space conversion method based on deep learning, and the method comprises the following steps: inputting video information into a feature extraction network, and obtaining multi-scale feature information of time and content information; inputting the multi-scale feature information into a multi-scale leaf layer classification network, and classifying the extracted multi-scale feature representation by using the leaf layer classification network to ensure that local and global information of the image can be accurately captured on different scales; the correlation volume between the input frames is calculated through a full-pixel correlation network, and optical flow estimation is obtained through iterative updating; through comparative learning, the video conversion effect is optimized in time and space dimensions; the overall consistency of the generated video sequence is ensured through cyclic consistency optimization; generating a target image by using a generator, and evaluating the quality of the generated image through a discriminator; and the generated high-quality image is converted into a video to be output.

Methods and apparatuses of encoding a video stream encoded using video point cloud coding include obtaining a source point cloud; generating an occupancy map including one or more pixels based on the source point cloud, an occupancy value associated with each pixel being a non-binary value; encoding the occupancy map to generate an encoded occupancy map, wherein a block of the encoded occupancy map corresponds to the one or more pixels of the occupancy map; selecting an occupancy value of the block based on the one or more pixels; and generating an encoded video stream based on the selected occupancy value.

In various examples, epipolar constraint-based cross-camera calibration validation is disclosed. For a pair of cameras that have partially overlapping fields of view, a shared region of their overlapping fields of view may be extracted and used as the basis to perform an epipolar constraint-guided feature descriptor matching process. A camera calibration metric may be computed based on the degree to which a feature descriptor appearing at a pixel of the first image aligns as expected in the second image with an epipolar line associated with the pixel of the first image, where the epipolar line is computed using extrinsic camera calibration parameters associated with the pair of cameras. Epipolar matching may be performed for a plurality of feature points and an aggregate validation score computed based on measuring the computed deviations for each feature. A sensitivity analysis may be applied to better assess the usefulness of the validation score.

A depth mapping apparatus includes a stereoscopic depth mapping module, including a first and second cameras configured to capture pairs of respective first and second images of a scene, and a range-sensing module, configured to measure respective ranges from the apparatus to multiple points in the scene. A controller processes a first pair of the first and second images of the scene to compute a first depth map of the scene and associates at least some of the points at which the range-sensing module measured respective ranges with corresponding pixels in the first depth map. The controller computes a correction function for the stereoscopic depth mapping module by comparing the respective ranges measured by the range-sensing module to respective depth coordinates of the corresponding pixels in the first depth map and applies the correction function in computing subsequent depth maps based on subsequent pairs of the first and second images.

A method for controlling a furnace for reheating iron and steel products, comprising forming an infrared image, using an infrared camera, of an upper face of a product over the width and at least partially over the length thereof when said product is arranged on a predetermined discharging surface; digital processing comprising binarization of the infrared image into two classes of pixels, one class that corresponds to the pixels associated with the presence of scale that is bonded on the face of the product and one class that corresponds to the pixels associated with the presence of scale that is not bonded on the face of the product; determining the amounts of non-bonded scale and of bonded scale on the upper face of the product on the basis of the binarized image; modifying furnace control parameters on the basis of the determined amounts of non-bonded scale and of bonded scale.

A multiview display includes an array of light valves having rows of a repeating plurality of color sub-pixels and arranged as a plurality of multiview pixels configured to modulate directional light beams as color pixels of views of a multiview image. A first row of the repeating plurality of color sub-pixels is offset from a second row of the repeating plurality of color sub-pixels in a row direction by an integer multiple of a width of a color sub-pixel. The offset of the first and second rows is configured to provide corresponding color sub-pixels in adjacent multiview pixels having different colors to mitigate color fringing associated with the color pixel of the multiview image. The multiview display further includes a light control film configured to control a view angle of the multiview image.

Processing circuitry (e.g., GPU) obtains a first MSAA intermediate buffer from a rendering of a first image, the first MSAA intermediate buffer includes a plurality of subsamples for each pixel of the first image. The processing circuitry determines, for a first pixel of the first image, a first surrounding subsample pattern based on first subsamples in a surrounding region of the first pixel. Further, the processing circuitry obtains, based on the first surrounding subsample pattern, a first blending weight vector for a first location among a plurality of locations that define output pixel locations within the first pixel for upscaling the first pixel to a plurality of output pixels in an upscaled image of higher resolution than the first image. The first blending weight vector includes first blending weights respectively for first candidate subsamples associated with the first pixel. The processing circuitry generates the upscaled image accordingly.

The present disclosure provides methods and systems for fusing chroma intra prediction modes. An exemplary method includes: generating a plurality of predicted chroma samples associated with a pixel, by using a plurality of chroma intra prediction modes respectively; and determining a first predicted chroma sample, based on a weighted sum of the plurality of predicted chroma samples.

A method of training a machine learning model to identify image features including providing training image data including pixels, assigning a groundtruth annotation to each pixel relating to a respective pixel and each groundtruth annotation indicating whether or not that the pixel corresponds with an image feature, providing an ignore mask including a set of ignore flags relating to a respective pixel and each ignore flag providing an indication that the pixel should be ignored, for each pixel, receiving a prediction value from the machine learning model indicating a probability of the pixel corresponding with an image feature, for each pixel which has no ignore flag, determining a loss value based on the prediction value and groundtruth annotation for that pixel, and training the machine learning model based on the loss value, and for each pixel having an ignore flag, ignoring the prediction value for that pixel.

Semantic segmentation generally refers to a machine learning process that associates a label or category with every pixel in an image. This can be used to recognize a collection of pixels that form distinct categories of objects, which may have applications in autonomous driving for example where the vehicle needs to identify other vehicles, pedestrians, traffic signs, pavement, and other road features from captured images of a surrounding environment. While using pixel-level annotations may be ideal for fully-supervised training of the semantic segmentation model, collecting such annotations is time-consuming and expensive, and therefore limits the scalability and practicality of fully-supervised training methods. The present disclosure enables weakly supervised training of a semantic segmentation model aided by learned prompts embedded with semantic knowledge.

A generation device (10) includes: a 3D reconstruction unit (12) that reconstructs an original three-dimensional image based on a plurality of images and a plurality of depth images, and acquires information indicating a position, a posture, a shape, and an appearance of a mapping target object that is a target of mapping to a digital space, and position information and posture information of an imaging device that has captured the image and the depth image; a labeling unit (13) that acquires a plurality of two-dimensional images in which labels or categories are associated with all pixels in an image based on the plurality of images; an estimation unit (14) that estimates a material and mass of the mapping target object based on the plurality of two-dimensional images and the position information and the posture information of the imaging device; and a generation unit (16) that integrates the information indicating a position, a posture, a shape, and an appearance of the mapping target object, and the information indicating the material and mass of the mapping target object, and generates digital twin data including position information, posture information, shape information, appearance information, material information, and mass information of the mapping target object.

A method for video encoding, a method for video decoding, apparatuses and non-transitory computer-readable storage media thereof are provided. In the method for video decoding, the decoder may obtain one or more filtering parameters, wherein the one or more filtering parameters are from a filtering parameter set comprising a number of similar blocks, a search range, and a window size; obtain, based on the one or more filtering parameters, one or more similar pixels for a first pixel that has been filtered in a picture and the first pixel; and obtain an original pixel associated with the first pixel based on the one or more similar pixels.