25 results about "Pixel mapping" patented technology

The application discloses a kind of fusion navigation information's heterogeneous scene matching point feature initial screening method, comprising: obtaining real-time graph, and the coarse matching of feature point is carried out to real-time graph and reference graph;For the first gradient amplitude chart corresponding to the modified real-time graph and the reference graph, and the first gradient direction chart corresponding is calculated;The first gradient direction chart is respectively mapped pixel by pixel;For real-time graph, determine radius based on the radius parameter and radius coefficient of pre-set;For the position coordinates of feature point on the modified real-time graph, construct histogram;Determine the main direction of each feature point, construct three regional main direction array, and determine the unique main direction array of feature point in real-time graph, reference graph;Determine the absolute value of main direction difference value;By the absolute value of main direction difference value and the matching effect index of pre-set, index array is constructed and is sorted, and the corresponding feature point pair is selected according to the number of required feature point pair.

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.

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.

The display method and display device for improving line sharpness are provided, the line is a red line in a red and black interval line or a blue line in a blue and black interval line, the display method comprises the following steps: obtaining a to-be-displayed original image of a display screen, the display screen is an active matrix organic light emitting diode display screen in which sub-pixels are arranged in a diamond arrangement manner; when performing sub-pixel rendering processing on the to-be-displayed original image, mapping a pixel in the line in the to-be-displayed original image to a sub-pixel with the same color as the pixel in a corresponding range of the pixel on the display screen, and configuring a non-zero pixel value for other sub-pixels in the corresponding range; and displaying sub-pixel data obtained after the to-be-displayed original image is subjected to the sub-pixel rendering processing through the display screen. The display method and display device can solve the problem that a red and black interval line or a blue and black interval line is not sharp enough when displayed on an AMOLED display screen arranged in a diamond arrangement manner.

The application provides an aerial advertising system and method with a flexible LED display screen carried by a UAV, and relates to the technical field of UAV display, which comprises the following steps: generating formation control instructions and task allocation information based on task requirements and preset advertising picture parameters; controlling the UAV cluster to ascend asynchronously in stages according to the formation control instructions and the task allocation information, completing the division of sub-clusters and hierarchical formation calibration in a target airspace, and forming an initial cluster formation; generating attitude adjustment instructions and display splicing calibration instructions based on the real-time spatial pose of each UAV in the initial cluster formation, and constructing a complete advertising display surface; performing perspective adaptation processing and partition pixel mapping on original advertising data according to the spatial topological characteristics of the complete advertising display surface, and generating adapted target video frame data; generating display trigger instructions based on a high-precision synchronous clock reference and a UAV display delay prediction result, and controlling each flexible LED display screen to synchronously play the target video frame data.

Disclosed are an image processing method and an image processing apparatus. An original image including a plurality of pixels is obtained. The plurality of pixels include a first pixel. N target pixel coordinates are recorded. The N target pixel coordinates form M target coordinate sets. M and N are integers greater than 0. The first pixel is associated with a first target coordinate set among the M target coordinate sets by utilizing a first classification method according to an original pixel coordinate of the first pixel. The first pixel is mapped to one of the N target pixel coordinates in the first target coordinate set by utilizing a second classification method. The original pixel coordinate of the first pixel is replaced with one of the N target pixel coordinates to convert the original image into an adjusted image.

This application relates to the field of image processing technology, and provides an image correction method, apparatus, device, and storage medium based on polarization features. The method acquires a main image and multi-polarization angle polarized images, and establishes a pixel mapping relationship between the main image and the polarized images, achieving precise spatial alignment of polarization features with the main image and obtaining the corresponding polarization features of the main image. Highlight regions are selected from the main image using preset evaluation conditions, and the polarization features are used to accurately identify specular reflection highlights, improving the recognition accuracy of highlight regions in the main image. The reflection characteristic components of each highlight region are determined based on the polarization features, and the highlight regions in the main image are corrected using these reflection characteristic components as physical constraints. This results in a corrected main image with more realistic colors, more natural transitions, and better visual consistency, thus improving the image correction quality.

Embodiments of the present application provide a light field 3D splicing view area correction method, device and related equipment, wherein the method comprises: dividing a plurality of target view areas and determining the corresponding standing probability of each target view area according to target scene information and related information of a plurality of target splicing screens; performing corresponding correction operations on the plurality of target view areas to generate a corrected view point pixel mapping sequence; determining a target view point pixel mapping sequence according to the view point pixel mapping sequence and the standing probability; and performing a target light field 3D splicing view area correction operation according to the target view point pixel mapping sequence. The above method can compensate for multi-dimensional errors in the light field 3D splicing display process, thereby reducing the deviation between the ideal light path and the actual light path, avoiding light crosstalk between different view points, and further avoiding problems such as misalignment, blurring and discontinuous view areas in the light field image viewed by the viewer through the splicing screen, so as to improve the light field image splicing quality.

This invention discloses a method and system for restoring polarized images of scattering environments based on manifold-constrained transmittance refinement. First, polarized images are acquired using a focal plane camera and channel compensation is performed to calculate Stokes parameters. Then, background polarization characteristics and background radiation are adaptively estimated. Next, an optical thickness proxy is constructed using polarization contrast, and an initial transmittance field is generated through a parameterized attenuation law with confidence weighting. Then, pixels are mapped to a five-dimensional polarization feature space to construct a data-adaptive manifold, and a Charbonnier penalty term is used to regularize and optimize the transmittance, resulting in a refined result that preserves the boundaries. Finally, scene radiation is inverted based on a physical model, and a closed-loop optimization of global parameters is performed using an unsupervised particle swarm optimization algorithm to improve the restoration quality. Compared with existing technologies, this invention can stably and accurately estimate transmittance, effectively suppress noise and artifacts, and is suitable for strong scattering environments such as underwater and foggy conditions.

This invention provides a method and system for converting elliptical ring images, relating to the field of image processing technology. The method includes: determining the elliptical ring parameters of an input target image containing an elliptical ring; wherein the elliptical ring parameters include the center coordinates of the elliptical ring, the major semi-axis of the elliptical ring, the minor semi-axis of the elliptical ring, the rotation angle, the starting angle, and the angle range; constructing an arc length-angle lookup table for the elliptical ring based on the elliptical ring parameters; determining the target angle corresponding to each rectangular column based on multiple rectangular columns contained in a preset rectangular image, the elliptical ring parameters, and the arc length-angle lookup table; determining the angular radial line corresponding to each target angle based on the rotation angle and the elliptical ring parameters; and sequentially mapping the pixels of the angular radial lines corresponding to each target angle to the rectangular image according to a preset direction to obtain the target rectangular image corresponding to the target image. This invention improves the accuracy of detection results for elliptical ring images.

This invention relates to the field of digital biomechanical assessment technology, providing a method for inverse calculation of dynamic pressure distribution, including: S1: offline initialization, performing finite element discretization on a transparent flexible elastomer, constructing and preprocessing a global stiffness matrix; S2: online scanning and displacement boundary mapping, acquiring the initial height field of the elastomer under no-load conditions and the real-time height field of the bottom of the test part after the elastomer is pressed under load; S3: multi-layer medium refraction dynamic compensation, performing layered refraction correction on the laser beam path based on the surface normal vector of the elastomer after compression; S4: inverse finite element analysis (iFEA) pressure inverse calculation, inputting the corrected displacement field as boundary conditions into a preset mechanical model; S5: geometric-dynamic data fusion output, mapping high-precision three-dimensional geometric mesh and high-resolution pressure texture pixel by pixel. This simultaneously acquires high-precision three-dimensional foot morphology and sub-millimeter-level pressure data, solving the problems of scanning distortion and disconnection between shape and pressure data.

The application discloses a kind of based on YUV color space's image fast splicing method, pixel mapping equation is established to the characteristic of the planar motion of driving mechanism, pixel mapping table is established in combination with the position of camera obtained by encoder and the image depth obtained by depth camera, so that the fast transformation of image pixels between different visual angles is realized;Pixel mapping equation is extended from a single pixel to matrix form, so that the whole picture or multiple pictures can be mapped in parallel, the fast splicing of large scene and multiple pictures can be realized, in the image splicing process, for the problem that there may be a seam in image fusion by directly using weighted average, while ensuring the image splicing effect, the time of image splicing is shortened.

This invention discloses a method, system, and medium for human fall detection based on multi-view geometry. The method includes: acquiring intrinsic and extrinsic parameters of multiple fixed cameras and a ground homography matrix; simultaneously acquiring images and obtaining a binary mask of the human body with global identifiers; determining radial lines based on the nadir points of each camera, adaptively selecting the optimal viewing angle camera, extracting foot and head image points, and calculating the horizontal position and vertical height using the homography matrix and geometric relationships; using multi-view ground homography constraints, mapping mask pixels in a reference view to other views for consistency verification, backprojecting verified pixels onto the ground plane to reconstruct the contact area and calculate the area; and fusing at least the temporal changes in height and contact area to determine whether a fall has occurred. This invention can accurately measure key physical quantities related to falls and has advantages such as strong robustness, high accuracy, good interpretability, and controllable hardware costs.

The application relates to the technical field of image recognition, and discloses a hollow capsule surface defect detection method based on visual detection. A capsule first image under a bright-field diffuse reflection light condition and a capsule second image under a dark-field directional side light condition are acquired; the first image and the second image are input into a pixel-level feature decoupling network, a capsule surface optical reflection model is taken as a physical constraint, and transmission light features in the first image and mirror reflection features in the second image are extracted; surface defect candidate pixels are mapped back to a spatial coordinate system of the first image through a homography matrix, and a detection result image is output through a decoder. The application binds an image recognition algorithm with capsule optical physical characteristics, excludes gray interference of transmission light and mirror reflection on defect features, and solves the technical problem of confusion of internal optical artifacts and surface physical defect features.

A headlamp system is provided for vehicles with an imaging unit and with an optical unit. The optical unit generates a light distribution that features a plurality of light patches. The light patches are each generated by mapping of at least one light pixel of the imaging unit. An actuator unit includes a corrective mechanism, by which, in event of presence of a defective light pixel that cannot be mapped by the optical unit on a light patch, at least one corrective light pixel is actuated, by means of which a light patch adjacent to the defective light patch not illuminated by the defective light pixel features a changed intensity progression in comparison to non-defective state. At least one flight patch adjacent to the defective light patch features an increased corrective intensity progression in comparison to the non-defective state.

The application discloses a hyperspectral image efficient splicing system based on grid guidance and manual registration, relates to computer vision technology, and is proposed in view of problems such as complicated operation in the prior art. The following steps are performed during running: a hyperspectral data folder is loaded and a false color image is generated; initial image splicing is performed; a coordinate mapping file is generated and saved; a global false color image is reconstructed based on the coordinate mapping file; and a single-band gray image is generated and interactive spectrum extraction is performed. The application has the advantages that an ordered splicing mechanism of grid arrangement guidance is provided, an accurate registration strategy of manual feature point selection is combined, and independent storage and on-demand reuse of a position mapping file are deeply integrated, so that decoupling scheduling of original hyperspectral data, a false color image and a pixel mapping relationship after initial splicing is realized, splicing accuracy and operation efficiency in a complex scene are significantly improved, memory occupation of secondary reconstruction and single-band spectrum analysis is reduced, and data reuse and interactive analysis capability are improved.

This application provides a vehicle-mounted projection system, method, and related apparatus based on high-speed UART. The vehicle-mounted projection system includes a System-on-a-Chip (SOC) and a projection lamp. The SOC includes a hardware acceleration module, a CPU, and a Vo module. The hardware acceleration module is used to decode and transform the video to be displayed, outputting grayscale pixels. The CPU receives the grayscale pixel data output by the hardware acceleration module, completing projection frame construction, UART frame encapsulation, and RGB pixel mapping. The Vo module is used to read and convert RGB pixels into corresponding level signals for output. The projection lamp is used to receive, parse, and finally project the level signals. This application fully releases the hardware potential of the SOC through the collaborative division of labor between the hardware acceleration module and the CPU, and overcomes the traditional UART transmission bottleneck by leveraging the high-speed output capability of the Vo module. This achieves efficient and stable vehicle-mounted projection while reducing system computing power and hardware costs.

An HD map system represents landmarks on a high definition map for autonomous vehicle navigation, including describing spatial location of lanes of a road and semantic information about each lane, and along with traffic signs and landmarks. The system generates lane lines designating lanes of roads based on, for example, mapping of camera image pixels with high probability of being on lane lines into a three-dimensional space, and locating / connecting center lines of the lane lines. The system builds a large connected network of lane elements and their connections as a lane element graph. The system also represents traffic signs based on camera images and detection and ranging sensor depth maps. These landmarks are used in building a high definition map that allows autonomous vehicles to safely navigate through their environments.

Embodiments of the present application provide a map generation method and device, electronic equipment and computer storage medium. The method comprises: obtaining a sequence of road images and a sequence of track points of a collection device of the road images, the track points being used to indicate the pose of the collection device in a world coordinate system when collecting the corresponding road image; obtaining a road surface area from the road image; obtaining a first spatial coordinate of a first pixel of the road surface area in a collection device coordinate system; determining a second spatial coordinate of a mapping point to which the first pixel is mapped in the world coordinate system according to the pose of the track point corresponding to the road surface area to which the first pixel belongs, so as to generate a point cloud map according to the second spatial coordinate. The method has better use effect.

The application provides an image reconstruction method and an electronic device, which can be applied to the technical field of medical image processing. The method comprises the following steps: performing registration on a plurality of pathological images based on a preset registration mode to obtain a plurality of target registration images; extracting a pixel mapping relationship between an mth target registration image and an ith pathological image by using a graph matching model to obtain an mth initial transformation matrix corresponding to the mth target registration image; performing image block cropping on the mth initial registration image and the ith pathological image by using a sliding window to obtain an mth initial registration image block and an ith pathological image block having a mapping relationship; performing non-rigid registration on the mth initial registration image block and the ith pathological image block by using an optical flow estimation model to obtain an mth reference registration image; determining an ith target registration image from m reference registration images related to the ith pathological image; and performing sequence mapping on the plurality of target registration images and the plurality of pathological images to obtain a reconstructed image.