121 results about "Pixel mapping" patented technology

The invention discloses a reinforcing steel bar surface binding point identification and positioning method based on point cloud and pixel mapping. The method comprises the following steps: obtaining original three-dimensional point cloud data to carry out outlier noise point elimination; performing plane fitting on the denoised point clouds, identifying template surface point clouds and removing the template surface point clouds; mapping the spatial physical coordinate attribute of the steel bar surface point cloud into a three-channel coding image; separating the three-channel coded image into three single-channel images: a length image, a width image and a depth image; processing the depth image to extract an initial reinforcing steel bar straight line set; screening out a parallel steel bar straight line set from the initial steel bar straight line set according to a set reference direction and a parallel angle tolerance threshold value; and calculating straight line intersection points in the parallel steel bar straight line set to obtain candidate binding point pixel coordinates, determining representative binding point pixel positions through clustering, querying coordinate information of the three single-channel images, and outputting three-dimensional space coordinates of the steel bar binding points. According to the invention, robust and accurate steel bar binding point identification and positioning are realized.

The invention relates to the technical field of machine vision detection and intelligent image processing, and discloses an intelligent track defect identification method based on machine vision. The track center is automatically positioned through a pixel coordinate system with the upper left corner as the original point in combination with row gray scale accumulation and maximum projection, the region of interest is dynamically intercepted, and background noise and calculation amount are effectively reduced. For the area, multi-angle equal-interval projection is adopted, pixels are mapped into a one-dimensional sequence, and the response to various strips and slender defects is improved. Each projection sequence is subjected to secondary trend fitting, a comprehensive residual image is constructed through the maximum residual, and defect signals are enhanced. According to the scheme, a threshold value is determined in a self-adaptive mode through statistics, four-neighborhood connected domain extraction and a dynamic area threshold value are combined, and a noise small domain is removed. Boundary, main direction, length and severity features are extracted from the remaining connected domains and mapped back to an original image, and efficient, accurate and automatic track defect intelligent recognition is achieved.

The invention relates to the technical field of endoscopes, and provides a spectral image processing method for a three-dimensional endoscope. The method comprises the following steps: synchronously acquiring single-band spectral images returned by multiple sensors of the endoscope; after the single-band spectral image is corrected, a two-dimensional spectral image is generated by combining multi-level fusion of the spectral weight; constructing a three-dimensional point cloud coordinate set through parallax calculation; mapping the two-dimensional spectrum fusion image to a three-dimensional point cloud based on a back projection sub-pixel mapping algorithm to obtain a three-dimensional spectrum point cloud; and performing Poisson fusion driven inter-block splicing fusion on the three-dimensional spectrum point cloud, and outputting an interactive three-dimensional navigation model. The technical problem that image details are lost and three-dimensional reconstruction is inaccurate due to the fact that spectral image fusion precision is insufficient in an existing three-dimensional endoscope image processing method is solved, registration and reconstruction of high-precision three-dimensional spectral point clouds are achieved through multi-modal image collaborative correction and optimization fusion, and the image fusion precision is improved. And the accuracy and the real-time performance of navigation in the endoscope are improved.

The invention relates to the technical field of target tracking, in particular to a panoramic target tracking method and device based on fusion positioning, equipment and a storage medium. Comprising the steps of obtaining a panoramic video stream and a positioning coordinate sequence; performing pixel mapping on each positioning coordinate in the positioning coordinate sequence to obtain a current mapping coordinate and a historical mapping coordinate; performing delay compensation on the current mapping coordinates based on the historical mapping coordinates to obtain predicted mapping coordinates; performing target detection on the panoramic video stream to obtain a visual coordinate set of the current frame, and performing spatial matching on the predicted mapping coordinate and each visual coordinate in the visual coordinate set to form a matching candidate set; if the matching candidate set is null, outputting the predicted mapping coordinate as a target position coordinate; and if the matching candidate set comprises at least one candidate visual coordinate, outputting the candidate visual coordinate closest to the predicted mapping coordinate as a target position coordinate. According to the invention, the tracking performance under a complex environment condition can be improved.

The invention relates to a forest region satellite fire point monitoring method based on deep learning, and the method comprises the steps: collecting a multispectral image of a target forest region through a satellite remote sensing system, and carrying out the preprocessing of the multispectral image; feature information extraction is carried out on the preprocessed multispectral image, the extracted feature information is classified by using a convolutional neural network, and a classification feature vector is obtained; fire source areas are determined according to the classification feature vectors, and a first fire source area set is obtained; acquiring environment data, and acquiring a second fire source region set through a support vector machine model according to the environment data and the preprocessed multispectral image; according to the first fire source area set and the second fire source area set, real fire point pixels are obtained; and real fire point pixels are mapped to geographic coordinates of the target forest region, missing data are filled by adopting a spatial interpolation technology, and final fire point position distribution is obtained. According to the method, the accuracy and reliability of forest fire detection under the complex terrain condition are remarkably improved.

The invention relates to a light effect animation data generation method and apparatus, a device and a medium. The method comprises the steps of obtaining a target image for constructing light effect animation data; according to the color values of the pixels in the target image, constructing a palette containing a fixed number of reference colors, and determining attribution probability distribution corresponding to each reference color mapped to the palette by each pixel; corresponding to the parity of each pixel in the plane space position of the target image, selecting a reference color according to the attribution probability distribution so as to update the color value of the pixel; and packaging the target image with the updated color value into the light effect animation data. According to the method, the light effect animation data is generated by utilizing the attribution probability and odd-even spatial modulation, and the smooth pixelated light effect can be realized when the atmosphere lamp is played.

The invention relates to the technical field of image detection, in particular to a power battery flatness detection method based on image recognition. The method comprises the following steps: acquiring a power battery image and calibration board information, and determining a pixel mapping relation; generating a battery surface point cloud picture by using the pixel mapping relation; fitting a reference plane by using the battery surface point cloud atlas, and calculating point cloud deviation; marking a battery bump area according to the point cloud deviation; calculating the height deviation of the bulge area of the battery, and determining the bulge degree; marking surface sunken points based on the battery surface point cloud atlas, and calculating the sunken depth of the surface sunken points; when the recess depth is greater than the recess threshold value, marking as a recess abnormal region; the flatness of the power battery is evaluated according to the bulging degree and the sunken abnormal area; and converting the flatness of the power battery into a three-dimensional rendering graph, and outputting a flat area grading result. According to the invention, accurate quantification of the battery flatness is realized based on an image detection technology, and the flatness detection accuracy is improved.

The invention relates to the technical field of image generation, in particular to a medical beauty 3D model construction method and system based on an image large model, and the method comprises the following steps: obtaining rasterized image data based on a 3D full-head point cloud model through space coordinate projection and pixel mapping, judging connectivity and a missing region, screening region features, and generating texture guidance parameters; and screening textures according to partition features, determining a multi-region fitting index, and outputting a continuous boundary fusion region. According to the invention, through carrying out multi-angle spatial reconstruction and fine pixel mapping on the three-dimensional point cloud data, texture coverage is not limited to a single view angle, spatial feature partitioning and target attribute identification, so that accurate supplementation of the texture of the missing region is realized; the locally generated texture content and the original input region are highly fused in the aspects of color structure and detail representation, and region mutation is effectively eliminated through spatial adaptation and boundary adaptive processing, smooth connection of multi-part texture splicing presentation, color transition and gradient fusion.

The invention provides a virtual camera projection method and system, electronic equipment and a medium, and belongs to the technical field of target detection, and the method comprises the steps: constructing a training data set according to sampling points in a three-dimensional space; modeling projection and back projection between the real camera view and the virtual camera view to obtain a two-stage multi-layer perceptron model; training a two-stage multi-layer perceptron model according to the training data set to obtain a trained model; and obtaining a pixel mapping table between the virtual camera view and the real camera view based on the trained model. According to the method, the reversible pixel mapping relation between the real camera and the virtual camera is constructed by adopting the two-stage multi-layer perceptron model, and the pictures under any group of camera parameters are uniformly converted to be under the virtual camera parameters, so that high-quality label data are efficiently generated through the automatic labeling model; the data utilization rate and the generalization ability of the automatic labeling model are remarkably improved, so that the model can directly process multi-source heterogeneous data, and the model adaptation cost is reduced.

A multi-camera imaging system includes a plurality of imaging units arranged side-by-side to capture images of a scene. A calibration module is configured to determine intrinsic and extrinsic parameters of camera modules of the imaging units and to establish a per-pixel mapping from image coordinates to a three-dimensional space. An image processing module selects, for individual of the plurality of imaging units, a field-of-view region within a captured image and blends the fields-of-view from adjacent imaging units together to form a composite image.

A multi-camera imaging system includes a plurality of imaging units arranged side-by-side to capture images of a scene. A calibration module is configured to determine intrinsic and extrinsic parameters of camera modules of the imaging units and to establish a per-pixel mapping from image coordinates to a three-dimensional space. An image processing module selects, for individual of the plurality of imaging units, a field-of-view region within a captured image and blends the fields-of-view from adjacent imaging units together to form a composite image.

The invention relates to the technical field of optical waveguide preparation, and discloses an optical waveguide mother set splicing-free preparation method based on a DMD dynamic mask, and the method comprises the steps: obtaining the layout data of an optical waveguide mother set as initial design information, and generating a mask pattern sequence which comprises a plurality of partition mask patterns; establishing a pixel mapping relationship between the on-off state of the micro-mirror unit and the partition mask pattern; driving the transparent substrate to execute partition positioning movement on the platform, and controlling the current exposure area of the photosensitive material layer to be aligned with the projection area of the partition mask pattern; on the basis of the pixel mapping relation and the partition positioning movement result, activating the mask pattern, and carrying out non-interval continuous exposure to form a full-page latent image pattern; the whole latent image pattern is subjected to developing treatment, the embossment structure of the optical waveguide master mask is treated and transferred into the transparent substrate, a waveguide core layer pattern is formed, the effects of non-interval continuous exposure and no segmented splicing are achieved, the preparation period of the surface embossment optical waveguide master mask is shortened, and the processing efficiency and the batch preparation yield are improved.

The application relates to the technical field of display, and particularly discloses an extended reality head-mounted display system based on Micro LED display, which comprises an eye movement tracking device, an image rendering processing unit, a bionic pixel mapping driving unit, a Micro LED display panel and an event-driven control unit; the line-of-sight center position is acquired through eye movement tracking, non-uniform rendering and bionic pixel mapping are realized, full-resolution pixels are driven in the line-of-sight center area, the data amount is reduced by using a pixel merging algorithm in the peripheral area, and the refresh is triggered only in the picture change area in combination with an event-driven mechanism. The application significantly reduces the computing power and bandwidth demand, relieves visual fatigue and dizziness, simultaneously realizes dynamic power consumption optimization, and improves the immersive experience and equipment endurance.

According to one embodiment, a processing apparatus for accelerating convolution operations comprises a memory array in which a plurality of elementary processing unit are arranged in an array structure, and an operation unit configured to temporarily store feature-map data, divide the feature-map data into predetermined units along a channel-axis direction according to a pixel-first mapping method, and deliver the divided data to the memory array in a manner of shifting the data according to a position of a kernel.

This application relates to a method, apparatus, electronic device, and storage medium for calculating road surface height. The method includes: acquiring lane lines; mapping lane line pixels to a first coordinate system to generate first coordinates; acquiring the next time-of-flight (RTK) point; calculating the distance between the next time-of-flight RTK point and the point containing the first coordinates to generate a deflection distance value; acquiring camera height data; processing the deflection distance value and the camera height data; and outputting the road surface height value. The solution provided in this application can replace the traditional method of calculating using the current time-of-flight RTK point and camera height data, thereby improving the accuracy of road surface height and thus improving the accuracy of the three-dimensional spatial coordinates of lane lines.

The present application relates to the technical field of image rendering, in particular to a rendering method and system for a model image interface, comprising the following steps: obtaining a view angle parameter and generating a screen pixel mapping result set of a surface patch, calculating a screen occupation ratio of the surface patch, generating a screen change feature of the surface patch in combination with a frame difference, generating a rendering scheduling sequence set based on the occupation ratio and the change feature, calling data in sequence to perform rasterization processing, and generating a rendering result of the model image interface. In the present application, the visual presentation weight under the current visual field is calculated by extracting vertex coordinate mapping, the difference in occupation ratio between the front and back frames is analyzed by using the finite difference method to capture the dynamic change trend, the dynamic rendering scheduling sequence is constructed based on the double dimensions of the occupation ratio and the change feature, the calculation resources are accurately tilted to the high screen occupation ratio and the high dynamic change area, the rendering quality of the core visual area is ensured, the invalid calculation overhead is greatly reduced, and the interface rendering real-time performance and frame rate stability in complex scenes are effectively improved.

The application provides a video blind area compensation transmission method and server based on dynamic field of view fusion, through acquiring multi-source video stream data, integrating all field of view regions under the same timestamp to a unified global reference space to obtain a global field of view splicing layout, constructing a dynamic field of view topology evolution graph containing a blind area hole outline and a pixel mapping transformation relationship between fields of view according to layout changes of continuous timestamps; performing pixel-by-pixel spatial migration on adjacent field of view video frames according to the pixel mapping transformation relationship of the blind area hole periphery to generate a blind area compensation video segment; performing panorama fusion on the compensation segment and the original video segment to obtain a panoramic video stream; analyzing the distribution pattern of the original collection source and the content remapping source in the panoramic video stream to generate a spatial mask, setting a differentiated encoding transmission strategy for different regions and encoding and distributing. The application realizes high-quality compensation of blind area content and regional differentiated efficient transmission of panoramic video streams.

This invention relates to the field of infrared imaging and low-level hardware processing technology, and discloses a parallel processing method and circuit for infrared image data based on FPGA. The method includes: establishing a pixel spatiotemporal mapping coordinate system and extracting pixel mapping point coordinate pairs; performing hardware topology sensing and spatial decoupling to output a spatial topology matrix; performing real-time pixel-level non-uniformity correction; implementing phase-locked noise reduction through a virtual offset field to obtain an equalized corrected bitstream; generating a quantization mapping curve based on saliency sensing; and performing real-time video reconstruction combined with adaptive power consumption control to output a video signal. This invention solves the problems of latency and instantaneous heat accumulation in large-area infrared data processing, achieving high-fidelity restoration of spatial topology and suppression of non-uniform noise; closed-loop energy management eliminates image blurring caused by thermal drift, comprehensively improving the system's imaging clarity and hardware operating efficiency.

A three-dimensional (3D) display monitor is disclosed, featuring an integrated processing chip designed to enhance real-time rendering and reduce system latency. The chip receives raw 3D image data and face tracking images, performs real-time face tracking and pixel mapping, and outputs the mapped image for display. The chip may receive direct eye center position data, bypassing face tracking images. The monitor includes functionalities including touch panel feedback, hand tracking, and automatic detection of 2D or 3D content, enabling seamless mode switching. Furthermore, the chip can convert 2D content to 3D content in real-time using a neural processing unit.

The present application relates to the technical fields of machine vision detection and intelligent image processing, and discloses a track defect intelligent identification method based on machine vision. Through the pixel coordinate system with the upper left corner as the origin, combined with the line gray accumulation and the maximum projection automatic positioning track center, the interested region is dynamically intercepted, and the background noise and the calculation amount are effectively reduced. For the region, multi-angle equidistant projection is adopted, the pixels are mapped into one-dimensional sequence, and the response to various types of strips and slender defects is improved. Each projection sequence is fitted by quadratic trend, the maximum residual error is used to construct a comprehensive residual image, and the defect signal is strengthened. The scheme uses statistical quantity to adaptively determine the threshold, combines the four-neighbor connected domain extraction and the dynamic area threshold, and removes the small noise domain. The boundary, main direction, length and severity of the remaining connected domain are extracted and mapped back to the original image, so that efficient, accurate and automatic track defect intelligent identification is realized.

The application discloses a kind of self-adapting projection methods based on high light boundary three-dimensional remapping.First, based on phase shift brightness statistics diagram, extract single or multiple overexposure connected region, extract the pixel group coordinates of the un-overexposed boundary of the region and reconstruct the corresponding three-dimensional coordinates of boundary based on calibration parameter;Then, based on the projection light path model, the three-dimensional coordinates are inversely mapped to the DMD target surface to obtain a closed-loop connected domain;High light closed-loop connected domain is screened using a two-dimensional image to obtain the boundary of the outermost circle high brightness range, and the projection light intensity is gradually reduced by high light boundary extraction and three-dimensional inverse mapping, which reduces the high light range, and finally completely eliminates the overexposure range.The method first maps the high light edge pixels to the DMD target surface pixels, and reduces the projection light to eliminate all high light areas.The complex splicing process of different three-dimensional point clouds in the traditional multi-exposure method is avoided, which improves the reconstruction rate while reducing the complexity of calculation.

The invention relates to the technical field of medical image processing, in particular to a kidney ultrasonic image-based fibrosis quantitative evaluation method and system, and the method comprises the steps: obtaining an ultrasonic image, carrying out the standardization processing, and estimating a system response correction function and an optical flow field; selecting a key frame based on an optical flow field, performing registration and constructing pose map optimization, fusing a envelope to establish mapping between a cortical conformal coordinate system and a key frame pixel, and generating a cortical region probability field and a boundary uncertainty field; calculating a sound attenuation coefficient and a scattering statistical parameter in the sliding analysis window, correcting the sound attenuation coefficient and the scattering statistical parameter, projecting the sound attenuation coefficient and the scattering statistical parameter to a conformal coordinate system, carrying out Wasserstein gravity center fusion according to probability and uncertainty weighting, detecting the effective thickness of the cortex, and carrying out consensus optimization to obtain a fibrosis fraction field and a consensus residual field; and generating a non-consistency total score, outputting a confidence interval and a confidence level through conformal prediction calibration, and generating a fallback recalculation or re-collection instruction if necessary. The cross-view, cross-time and cross-equipment consistency is improved, and the closed-loop quality control capability is achieved.

A method for power-saving in a user equipment (UE). The method may include identifying user interface (UI) elements or regions of interest (ROIs) present on a display screen of the UE. The method may include obtaining a pixel map and generating a binary mask, based on mapping pixels in the pixel map with the identified at least one of UI elements or ROIs. Further, the method may include adjusting the brightness of pixels outside of the identified at least one of UI elements or the ROIs, thereby saving power in the UE using adaptive brightness.

An electronic device may include a lenticular display. The lenticular display may have a lenticular lens film formed over an array of pixels. The lenticular lenses may be configured to enable stereoscopic viewing of the display such that a viewer perceives three-dimensional images. Display pipeline circuitry for the lenticular display may implement crosstalk correction. Crosstalk correction may be performed before pixel mapping using stored crosstalk calibration profiles, crosstalk correction may be performed by ray tracing multiple beams per pixel, and / or crosstalk correction may be performed after pixel mapping using a kernel lookup table.

The invention discloses a cultural relic disease intelligent diagnosis system based on multispectral fusion and AI, and the system comprises a multispectral data collection module, a spectral feature extraction module, a cross-modal data association module, an AI disease recognition and analysis module, a diagnosis result generation module, and a system interaction control module. Transmitting to a spectral feature extraction module to extract spectral features; the cross-modal data association module establishes spectral feature and cultural relic image pixel mapping and generates an association data set; the AI disease recognition and analysis module processes the data set and recognizes disease types and region boundaries; the diagnosis result generation module integrates the information to generate a report containing disease position, area and severity; and the system interaction control module displays the report and realizes storage and external calling. Through multi-module cooperation, multispectral and AI technologies are fused, the historical relic disease diagnosis precision and efficiency are improved, and comprehensive data support is provided for historical relic protection.

The invention discloses a dynamic switching method and device based on EDID identification, and the method comprises the steps: receiving an eDP signal through an eDP input interface, and dividing the eDP signal into two paths through an eDP signal distribution unit; reading the EDID information from a display device connected with the first output interface and the second output interface through the control unit; the interface type, the optimal resolution, the time sequence format and the pixel mapping format supported by each display device are judged through the EDID information analyzed by the control unit; generating a control signal through a control unit according to the judgment result; and according to the configured mode, the two paths of signals are processed and converted through the corresponding signal processing paths respectively, and are finally output to display equipment through the corresponding output interfaces. According to the method and the device, plug and play is realized, manual intervention of a user is not needed, and the optimal display mode is automatically adapted. The device can be adapted to display devices with different interfaces, different resolutions and different brands, and the product applicability is greatly improved.

The invention discloses an optical window distortion correction method based on static contrast calibration and optical angle deviation. The method comprises the following steps of: a, obtaining a pixel-by-pixel light window distortion mapping pixel deviation table Tc and a pixel-by-pixel mask table Tm through static comparison calibration; b, obtaining an external parameter matrix of a camera imaging origin and a light window under a body coordinate system based on real external parameter calibration of a total station camera; c, performing optical simulation analysis by taking a field angle parameter of the camera as an input parameter to obtain an original light incidence vector with optical angle deviation and a deviated light incidence vector; d, performing distortion mapping on the original light incidence vector with the optical angle deviation and the deviated light incidence vector to obtain a distortion mapping pixel deviation table T1 caused by the optical angle deviation; and e, fusing the Tc, the Tm and the T1 to obtain a distortion pixel mapping table T0 of the full picture pixel by pixel. According to the invention, precise distortion separation and high-precision correction can be realized.

Embodiments perform gradient-based interpolating pixels. In one embodiment, a method comprises: converting point cloud data into one or more range images, wherein each pixel in a range image within the one or more range images is to map to a point represented by a set of cartesian coordinates in the point cloud; dividing a range image within the one or more range images into a plurality of windows, each window including a set of pixels in the range image; adding one or more pixels to at least one window through interpolation of pixels within the window, wherein one or both gradients between a first pixel and the first pixel's two immediately adjacent pixels in the window are used to determine a first interpolated pixel to be added; and converting pixels within the plurality of windows with one or more interpolated pixels into updated point cloud data of the point cloud.

The invention provides a high-speed railway multi-professional data unified registration method and device based on visual features, and relates to the technical field of railway infrastructure operation, maintenance and detection, and the method comprises the steps: obtaining multi-source data; respectively extracting facility feature points from the geometric one-dimensional data of the track and the one-dimensional data of the overhead line system, and outputting the category and mileage position of the facility through identification processing; image facility feature points are detected from line-scan digital camera image data, facility categories and image pixel coordinates are output through detection, and a unified coordinate system is constructed based on the image pixel coordinates; grouping according to facility types, and matching the track inspection facility feature points with the image facility points and matching the network inspection facility points with the image facility points to obtain matching results; and piecewise linear spreading is carried out between adjacent matching pairs, a monotonous mileage-to-image pixel mapping function is generated according to a mileage-to-image pixel mapping relation, multi-source data is mapped to a unified coordinate system, and a registration result is obtained.