623 results about "Pixel pair" patented technology

The invention relates to a circuit board defect identification method and system based on multi-dimensional image data, and belongs to the technical field of data identification processing, and the method comprises the following steps: obtaining synchronous image data of a circuit board to be detected in a plurality of imaging modes; performing space-spectrum joint registration on each modal image to generate a multi-dimensional image cube with a unified coordinate system and pixel alignment; inputting the multi-dimensional image cube into a pre-trained multi-branch heterogeneous fusion neural network; generating a pixel-level defect probability graph by utilizing a defect sensing context decoder, and performing geometric constraint optimization on the probability graph by combining prior information of a circuit board design layout; outputting defect types, positions and confidence coefficients, and establishing an interpretable defect fingerprint database according to the multi-dimensional response characteristics of the defects; the method has the beneficial effects that false defect signals generated by image noise and circuit board surface texture interference can be effectively inhibited, the omission ratio and the false detection ratio are greatly reduced, and pixel-level accurate defect positioning is realized.

A computing system for generating a gain map-based enhanced High Dynamic Resolution (HDR) image receives an HDR image comprising a plurality of pixels, each pixel having one or a plurality of brightness values for each of a plurality of color components, applies at least a first tone mapping algorithm to each pixel in the HDR image, generates a gain map using the HDR image and the SDR image, generates a gain map-based enhanced HDR image based on the SDR image and the gain map, and generates an output based on the gain map-based enhanced HDR image.

The invention belongs to the technical field of remote sensing image processing, computer vision and deep learning, and particularly relates to a remote sensing image erosion gully semantic segmentation method based on improved OfficientNet-UNet. Comprising the following steps of 1, data preparation and data preprocessing; 2, constructing and enhancing a data set; step 3, model construction and strategy training; and 4, performing contrast experiment and result evaluation. According to the method, detail features of ground features can be more accurately captured, an overfitting phenomenon caused by too high model complexity is reduced, so that the classification precision and boundary recognition accuracy of land coverage data are effectively improved, weights of different types of samples can be automatically adjusted in the training process, and the training efficiency is improved. Particularly, the contribution of background pixels to a loss function is reduced, so that the problem of dominant training of the background pixels is effectively relieved; the method has good expansibility.

Provided is an EVS to which an image plane phase difference method AF function can be easily applied.A photodetection device includes: a first pixel configured to photoelectrically convert incident light from a portion on one side to generate a first current; a second pixel configured to photoelectrically convert the incident light from a portion on a side opposite to the portion of the first pixel to generate a second current; a first luminance circuit provided individually corresponding to the first pixel and configured to generate a first luminance signal according to the first current; a second luminance circuit provided individually corresponding to the second pixel and configured to generate a second luminance signal according to the second current; a first current comparison circuit provided individually corresponding to the first pixel, and configured to compare the first luminance signal with a reference signal and convert the first luminance signal into a first digital signal; a second current comparison circuit provided individually corresponding to the second pixel, and configured to compare the second luminance signal with the reference signal and convert the second luminance signal into a second digital signal; and a control unit configured to adjust a focal position of the incident light on the basis of the first digital signal and the second digital signal.

The invention relates to an unmanned aerial vehicle image target level change detection method based on multi-level fusion in the technical field of image processing. Registering two unmanned aerial vehicle images of the same scene and different time phases, and performing pixel alignment on an overlapping region of the two images according to registration points; processing the two images of which the pixels are aligned, and outputting a pixel-level change detection result; training a universal target detection model, and performing target detection on the two images with the pixels aligned by using the model; and fusing the target detection result with the pixel-level change detection result to obtain a target-level change detection result. According to the invention, an unmanned aerial vehicle image target level change detection task is decomposed into an unmanned aerial vehicle image pixel level change detection sub-task, a target detection sub-task and a multi-level fusion sub-task, so that the task difficulty is reduced, and the problems of insufficient training samples and poor generalization performance in the prior art are fundamentally solved; and the target level change detection speed is high, the accuracy is high, the application range is wide, and the robustness is high.

The present disclosure relates to a method of contrast-based autofocus of an image capture device, the method comprising: determining if at least one initiation condition is met, comprising: obtaining a first contrast metric representative of a contrast characteristic of at least one region of a first image captured with a first focus setting, the first contrast metric being determined based on one or more first data values corresponding to one or more pixels of the first image; determining a change in the contrast characteristic of the at least one region of the first image based on the first contrast metric with respect to a reference contrast metric obtained from a previous image captured with the first focus setting; determining if the change in the contrast characteristic equals to or exceeds an initiation threshold for the first focus setting, wherein the initiation threshold for the first focus setting is set based at least in part on the first focus setting; and upon determining that at least one initiation condition is met, initiating an autofocus process.

The invention belongs to the technical field of computer vision and digital media, and discloses a dome-screen film production method and system based on a high-altitude aerial image. Projecting the multi-view aerial video sequence to a unified spherical coordinate system, and constructing a panoramic space with an overlapping region; performing non-rigid geometric deformation processing on the overlapping region of the panoramic space to generate an intermediate image with aligned pixels; calculating the fusion cost of each pixel point in the overlapping region in the intermediate image, and performing path search based on the fusion cost of each pixel point to generate a dynamic suture line; generating an image mask based on the dynamic suture line, performing multi-scale frequency domain adaptive fusion on the intermediate image by using the image mask to generate a fused panoramic image, and remapping the fused panoramic image into a projection format adaptive to dome screen playing to output; and the real reducibility and the manufacturing quality of the dome film are greatly improved.

The invention discloses a brightness extraction method, electronic equipment, a storage medium and a program product, and relates to the technical field of near-to-eye display, and the method comprises the steps: when an instruction of carrying out the brightness extraction of a display screen of near-to-eye display equipment through a target camera is detected; the mapping relation between camera pixels of a target camera and screen pixels of the display screen is determined, and the target camera is located at an exit pupil frame of the near-eye display device; displaying the calibration brightness image on a display screen, and shooting through a target camera to obtain a calibration brightness camera image corresponding to the calibration brightness image; and based on the mapping relationship, extracting brightness information corresponding to each screen pixel on the display screen from the calibrated brightness camera image. According to the invention, under the condition that the camera image is seriously distorted due to distortion of the optical system, the accurate brightness information completely matched with the original pixel distribution of the screen can be reliably and efficiently obtained.

The invention discloses a single-frame unmanned aerial vehicle image pixel positioning method and system based on an elevation map, and relates to the technical field of unmanned aerial vehicle image processing and geographic positioning, and the method comprises the steps: obtaining a single-frame original image shot by an unmanned aerial vehicle, unmanned aerial vehicle attitude information during shooting, camera internal and external parameters, and a digital elevation map of a shooting region; according to the internal reference of the camera, converting the pixel coordinate of the target pixel in the original image into a normalized plane coordinate system of the camera, and carrying out coordinate distortion correction to obtain the back projection point coordinate of the target pixel; converting a camera-unmanned aerial vehicle body coordinate system and a body-geographic inertial reference coordinate system according to camera external parameters and unmanned aerial vehicle attitude information to obtain normalized ray direction vectors corresponding to target pixels; an earth surface height constraint is constructed based on a digital elevation map, and an intersection point of a ray and the ground is determined by performing iterative approximation on a space point along a ray direction. According to the invention, rapid and accurate conversion from the single-frame unmanned aerial vehicle image pixel to the geographic coordinate can be realized.

The invention provides a numerical control machining tool wear multispectral imaging detection method and system. The method comprises the steps that a three-dimensional data cube is constructed through a multispectral image sequence of a tool machining area; defining a spectrum detection window of each pixel point in the three-dimensional data cube, dividing the spectrum detection window into a main detection sub-block and auxiliary detection sub-blocks, and generating a spectrum difference feature map based on a difference relationship between a spectrum curve of each pixel in the main detection sub-block and an average spectrum curve of all the auxiliary detection sub-blocks; determining a projection vector according to a target spectral feature in the center detection sub-block and a background spectral feature in the local detection window; and determining a spectral feature map according to the projection vector and the three-dimensional data cube, and recognizing an abnormal wear area in the tool machining area according to a multidirectional gradient feature map constructed by a spatial gradient operator corresponding to each pixel in the spectral feature map and a spectral difference feature map. According to the technical scheme provided by the invention, the subtle difference between the wear region and the background region can be distinguished in a multi-dimensional interference state.

The invention provides a printing image processing method, a thermal printing control method and electronic equipment, and the method comprises the steps: extracting the page information of a to-be-printed document, and converting the to-be-printed document into a bitmap according to the page information; obtaining a global threshold value of each pixel point based on the bitmap; sequentially traversing each pixel in the bitmap, for the traversed target pixel, obtaining a region contrast ratio corresponding to the target pixel, and adjusting a global threshold value according to the region contrast ratio to obtain a local threshold value corresponding to the target pixel; and binarizing the corresponding target pixel according to each local threshold to obtain a binarization result. The local threshold is obtained by adjusting the global threshold through the area contrast, the local threshold can better reflect the pixel color close to the target pixel, the problem that the global threshold cannot effectively distinguish the black pixel from the white pixel in the partial area with the low contrast is solved, accurate binarization operation is performed on the target pixel, and the image quality is improved. The contrast between black pixels and white pixels is improved, and the black and white printing effect is improved.

The invention discloses a large-view-field camera light calibration method and system based on polynomial fitting, and the method comprises the steps: setting a measurement target on a camera and a calibration plate rigid tool, building a fixed connection coordinate system, directly measuring the pose between the coordinate systems through employing a laser tracker and other external high-precision equipment, and achieving the decoupling of a pose error and an image algorithm. The method comprises the following steps of: acquiring a plurality of calibration plate images, performing global polynomial curved surface fitting on sparse feature angular points of each image, and establishing continuous analytic mapping from pixels to a three-dimensional space; and carrying out spatial straight line fitting on a plurality of three-dimensional points corresponding to the same pixel to obtain an accurate light equation of the pixel. Calibration results are stored based on a camera target seat coordinate system, and direct migration and use can be achieved only by re-measuring the conversion relation between the coordinate system and a world coordinate system during actual measurement. The method is suitable for any complex imaging system, and has the advantages of high precision, strong universality, good robustness, outstanding engineering practicability and the like.

A distance measuring device that outputs one or more distance information items each indicating a distance to a subject, the distance measuring device includes: a light source that emits illumination light including one or more bright portions; a light receiver that includes a pixel array in which pixels are arranged in a matrix; a distance information calculator that calculates, for each of the pixels, pixel distance information indicating a distance from the pixel to the subject when the light receiver receives reflected light that is the illumination light reflected off the subject; a distance information selector that selects one or more selective distance information items from among the pixel distance information items corresponding to the pixels; and an outputter that outputs the one or more distance information items, based on the one or more selective distance information items.

The invention discloses a driving method for a display device. The display device comprises a plurality of sub-pixels, scanning lines and data lines. The method comprises the following steps of: receiving picture data, wherein the picture data comprises a plurality of sub-pixel gray-scale values to correspond to sub-pixels; the absolute value of the difference value between a first sub-pixel gray-scale value corresponding to a first sub-pixel and a second sub-pixel gray-scale value corresponding to a second sub-pixel is calculated, the first sub-pixel is electrically connected with one of a data line and two scanning lines, and the second sub-pixel is electrically connected with the other one of the data line and the two scanning lines; and when the absolute value of the difference value is greater than or equal to the first threshold value, adjusting the sub-pixel gray-scale value of at least one of the first sub-pixel and the second sub-pixel. Wherein a difference absolute value between the adjusted sub-pixel gray-scale value of the first sub-pixel and the adjusted sub-pixel gray-scale value of the second sub-pixel is smaller than a difference absolute value between the adjusted sub-pixel gray-scale value of the first sub-pixel and the adjusted sub-pixel gray-scale value of the second sub-pixel.

An imaging device is provided with: an imaging unit having a first pixel that photoelectrically converts light that has passed through an optical system and outputs a signal for focus detection, and a second pixel that photoelectrically converts light that has passed through the optical system and outputs a signal for image generation; a focus detection unit that performs focus detection of the optical system on the basis of a signal output from the first pixel; an evaluation unit that evaluates the contrast of the image on the basis of the signal output from the second pixel in a region including a subject in the image on the basis of the signal output from the second pixel; and a control unit that controls the focus detection by the focus detection unit on the basis of the evaluation by the evaluation unit.

An image processing method, an electronic device, and a storage medium are provided. The method includes: obtaining an image to be processed in response to an image processing request, wherein the image to be processed includes a first object; determining first distance values from a plurality of image pixels in the image to be processed to an object boundary of the first object, and determining a first pixel corresponding to the first object from the plurality of image pixels according to the first distance values; and determining a shadow pixel value for the first object acting on the first pixel according to a preset color, determining a first image according to the shadow pixel value corresponding to the first pixel, and displaying the first image.

The present application provides an information processing method, device and equipment, wherein the information processing method comprises: determining a target feature map corresponding to a target object according to an image size of the target object and first information; obtaining target coordinates of each pixel on the target feature map according to original coordinates of each pixel on the target mask and an input picture; obtaining a center coordinate of the target object according to the target coordinates; obtaining a target weight corresponding to each pixel according to the target coordinates and the center coordinate; training an object detection model according to the target weight corresponding to each pixel; wherein the first information comprises a mapping relationship between an object image size and a feature map on a feature pyramid; the target feature map is a feature map of a target layer on the feature pyramid; and the target object corresponds to the target mask. The present application improves the model precision obtained by the information processing scheme for training the target detection model in the prior art.

The application discloses a display device, a brightness compensation method, a storage medium and a computer device. The brightness compensation device of the display device of one embodiment comprises: a gray scale data acquisition module for acquiring gray scale data of each pixel, wherein the gray scale data comprises a pixel category and gray scale information; a distribution data calculation module for obtaining, from the gray scale data, a pixel category and a count value corresponding to the largest number of pixels at each gray scale value from among a first pixel, a second pixel and a third pixel, as distribution data; a reference data acquisition module for determining, from the distribution data corresponding to different gray scale values, distribution data with the largest count value as reference data; a brightness gain value determination module for determining, according to a preset gain model, an input parameter of the gain model from the gray scale data corresponding to the reference data, and obtaining a gain value; and a brightness compensation module for performing brightness compensation on a brightness value corresponding to each color pixel according to the brightness gain value.

This application discloses a method, apparatus, device, and storage medium for crosstalk compensation in a display panel. The method includes: determining a first typical value and a second typical value based on the luminance parameters corresponding to some luminous sub-pixels in every two adjacent pixel rows; ensuring that any one of the luminous sub-pixels corresponding to the first typical value is not located in the same luminous sub-pixel column as any one of the luminous sub-pixels corresponding to the second typical value; determining a first compensation parameter based on the luminance parameter compensation relationship, the first typical value, and the second typical value; and determining a final compensation value based on the first compensation parameter. According to embodiments of this application, the final compensation value for each pixel row can be determined based on the luminance difference of some luminous sub-pixels in every two adjacent pixel rows, thereby compensating and adjusting the data signal voltage of each row of luminous sub-pixels to improve the problem of signal crosstalk generated when the power signal line changes data voltage, and ensuring the display effect of the display panel.

Techniques are provided for facilitating detection threshold determination for infrared imaging systems and methods. In one example, a method includes capturing, by an imaging device, a thermal image of a scene. The method further includes determining temperature difference data indicative of a difference between temperature data of the thermal image associated with a background of the scene and temperature data of the thermal image associated with gas detection. The method further includes determining detection threshold data based on sensitivity characteristics associated with the imaging device and the temperature difference data. The method further includes generating a detection threshold image based on the detection threshold data. Each pixel of the detection threshold image corresponds to a respective pixel of the thermal image and has a value indicative of a detection threshold associated with the respective pixel of the thermal image. Related devices and systems are also provided.

The application discloses a 4D video generation method fusing monocular vision and terrain elevation data, comprising the following steps: acquiring video frame images and corresponding camera pose data collected by a monocular camera; acquiring terrain elevation data of a target area, and constructing a three-dimensional elevation network model; based on the camera pose data, establishing a ray projection model for pixels in the video frame images, calculating the intersection of the rays and the three-dimensional grid model, and obtaining the reference distance from the monocular camera to the ground surface corresponding to the pixels; generating an initial relative depth map of the video frame images by using a depth estimation model; selecting high-confidence ground pixels from the video frame images as reference points, taking the reference distance corresponding to the reference points as the true value, and performing absolute scale correction on the initial relative depth map to obtain an absolute depth map; performing back projection calculation on the absolute depth map, the camera pose data and pixel color information to generate a three-dimensional point cloud, mapping the three-dimensional point cloud to a real geographic coordinate system, and generating 4D video data in time sequence.

The application provides a target tracking method and system fusing co-occurrence statistics and fhog gradient features, and the method comprises the following steps: determining an initial position according to a target region; calculating a target position p t and a scale factor s t‑1 of a previous frame of a current frame I t‑1 , so as to determine a target region and extract relevant features of the current frame; co-occurrence filter is used to obtain co-occurrence statistical information between pixel pairs in a tracked target image, fDSST algorithm is used to obtain fhog features of the tracking target, and the co-occurrence statistical information and the fhog features of the tracking target are fused, so as to be used as target image features; when the model is updated, the target region is determined according to a target center p t and a target scale s t of the current frame I t , and the target region is extracted; the target region extraction features are sent into a position filter and a scale filter, so as to obtain a target tracking result through iterative updating. The application solves the technical problems that the prior art cannot completely and effectively use feature expression of a target, has low robustness, and has poor tracking effect in a specific scene.

A large field of view camera light ray calibration method and system based on polynomial fitting, the method is that the camera and the calibration board rigid tooling are respectively provided with measurement targets, a fixed connection coordinate system is established, and an external high-precision device such as a laser tracker is used to directly measure the pose between the coordinate systems, and the decoupling of the pose error and the image algorithm is realized. A plurality of calibration board images are collected, a global polynomial curved surface fitting is respectively performed on the sparse feature corner points of each image, a continuous analytical mapping of pixels to three-dimensional space is established, and then a plurality of three-dimensional points corresponding to the same pixel are subjected to spatial straight line fitting, so that the accurate light ray equation of the pixel is obtained. The calibration result is stored based on the camera target seat coordinate system, and when actually measuring, only the conversion relationship between the coordinate system and the world coordinate system needs to be measured again, so that the calibration result can be directly migrated and used. The application is suitable for any complex imaging system, and has the advantages of high precision, strong universality, good robustness and outstanding engineering practicability.

Provided is a light receiving element capable of enhancing accuracy of measurement of color mixing components contained in output signals of light shielding pixels located around a light receiving pixel. A peripheral pixel region has a specific pixel group including a light receiving pixel and light shielding pixels surrounding the light receiving pixel. A first conductivity type region corresponding to the light receiving pixel and the light shielding pixels of the specific pixel group is made lower in concentration of a first conductivity type impurity than a first conductivity type region corresponding to an effective pixel located in an effective pixel region. Since the concentration of the first conductivity type impurity of the specific pixel group (the light receiving pixel and the light shielding pixels) is low, it is possible to weaken a pn junction strength and suppress dark current in the light receiving pixel and the light shielding pixels. It is therefore possible to reduce noise contained in the output signals of the light shielding pixels and enhance the accuracy of measurement of the color mixing components included in the output signals of the light shielding pixels.

The application relates to a binocular image key point matching method based on a hierarchical optimization strategy and a medium, and the method comprises the following steps: S1, acquiring multiple pairs of left-right stereo image pairs; S2, processing the left-right stereo image pairs by using a deep neural network feature extractor to obtain left-right feature map pairs with different resolutions; S3, calculating matching cost volumes of the left-right feature map pairs to construct a matching cost volume pyramid; S4, obtaining an initial key point matching pair based on a matching cost volume with the lowest resolution; S5, calculating the matching cost between corresponding image blocks of the key point matching pair in a matching cost volume with a lower resolution, and taking a pixel pair corresponding to the matching cost satisfying local extremum as a key point matching result of the lowest level; and S6, repeating step S5, and performing layer-by-layer optimization on the matching cost volume pyramid in a manner from low to high resolutions until a final key point matching result is obtained. Compared with the prior art, the application has the advantages of high precision and strong robustness.

A display panel, a display apparatus, and an electronic device. The display panel comprises an array substrate (1) and a color filter substrate (2) which are stacked; the color filter substrate (2) comprises a color filter layer; the color filter layer is divided into a plurality of pixel units (21) that are periodically and repeatedly arranged; each pixel unit (21) comprises a red sub-pixel (211), a green sub-pixel (212), a purple sub-pixel (213), and a sky blue sub-pixel (214); the array substrate (1) is provided with first pixel driving structures corresponding to the sky blue sub-pixels (214) and second pixel driving structures corresponding to the purple sub-pixels (213); the first pixel driving structures are used for controlling the brightness of light in a first target band transmitted by the sky blue sub-pixels (214); and the second pixel driving structures are used for controlling the brightness of light in a second target band transmitted by the purple sub-pixels (213).

The application discloses a display device and a control method. For any first pixel in a liquid crystal layer, the first pixel comprises two types of sub-pixels, i.e., a first sub-pixel and a second sub-pixel. In a first sub-field, a color image can be displayed by using the first sub-pixel, and in a second sub-field, a monochrome image can be displayed by using the second sub-pixel. In this way, the image intensity is concentrated in a single sub-field, the sensitivity of human body to color separation can be reduced, the color separation can be reduced, the display quality of the image can be improved, and since the area corresponding to the first sub-pixel is not covered by a filter layer or is covered by a transparent filter layer, the light output by a backlight module can be directly emitted, the light intensity loss can be reduced, the light transmittance of the panel can be improved, the power consumption of the display device can be reduced, and the balance between light efficiency and image quality is achieved.

The invention discloses a transformer submersible robot image enhancement method, which comprises the following steps: S100, reading an original image shot by a camera to obtain an in-oil imaging original image shot by a robot in a transformer, and obtaining a gray scale corresponding to each pixel; s200, establishing a three-dimensional coordinate according to the coordinate distribution of different pixels in combination with the corresponding gray scale, so as to obtain a three-dimensional image corresponding to the pixel surface; s300, further obtaining an original image cumulative probability histogram based on statistics of the number of pixel points corresponding to the gray scale of each degree; s400, performing optimized gray value mapping on the basis of the original image cumulative probability histogram; s500, based on the optimized gray value mapping, obtaining an equalized image; and S600, obtaining a histogram based on the equalized image. In a dark environment in the transformer, the image shot by the robot can be enhanced, so that the robot can replace people to carry out transformer internal inspection.

The application discloses a kind of single frame unmanned plane image pixel positioning method and system based on elevation map, it is related to unmanned plane image processing and geographic positioning technical field, obtain the single frame original image of unmanned plane shooting and the digital elevation map of the shooting area when unmanned plane attitude information, camera internal and external parameters, shooting;According to camera internal parameter, the pixel coordinates of target pixel in original image are converted to camera normalization plane coordinate system, and target pixel back projection point coordinates are obtained after coordinate distortion correction;According to camera external parameter and unmanned plane attitude information, camera-unmanned plane body coordinate system, body-geographic inertial reference coordinate system conversion is carried out, and the normalized ray direction vector corresponding to target pixel is obtained;Based on digital elevation map, construct ground height constraint, by iterative approximation to space point along ray direction, determine the intersection of ray and ground.The application can realize the fast and accurate conversion of single frame unmanned plane image pixel to geographic coordinate.

A computer device includes at least one processor in communication with at least one memory device. The at least one processor is programmed to: a) receive, from the image capture device, a plurality of images for a continuous scan phase shift interferometry (PSI), wherein each image of the plurality of images includes a first plurality of pixels of a first item and a second plurality of pixels of a partial cavity; b) perform intensity scans of each pixel in the second plurality of pixels; c) identify zero transitions for each of the second plurality of pixels; and d) statistically evaluate the identified zero transitions by polynomial regression analysis.