93 results about "Image edge" patented technology

This application discloses a method, apparatus, device, and storage medium for edge detection of images, relating to the field of image edge detection technology. The method includes: acquiring an input image and performing structure-preserving filtering on the input image to obtain a smoothed image; calculating a first-order gradient magnitude map based on the smoothed image; generating an adaptive parameter mapping based on the statistical characteristics of the first-order gradient magnitude map; obtaining an edge magnitude map based on the adaptive parameter mapping; determining a global threshold based on the edge magnitude map, and setting a high threshold and a low threshold based on the global threshold; and performing double-threshold hysteresis connection processing on the edge magnitude map based on the high threshold and the low threshold to output a binary edge map. This method can improve the detection accuracy of images.

A method for segmented fitting of shield tunnel circumferential joints based on gradient value feature map density clustering includes the following steps: unfolding the tunnel's three-dimensional laser point cloud data to generate a two-dimensional grayscale image; calculating the vertical gradient values ​​of all pixels in the grayscale image, and generating a positive gradient value feature map A, a negative gradient value feature map B, and a comprehensive gradient value feature map C based on the magnitude and sign of the gradient values; determining the initial upper and lower edge row coordinates of the circumferential joint based on Figures A and B, and taking the middle row as the initial circumferential joint position; extracting pixels from the circumferential joint region in Figure C based on the initial circumferential joint position, performing cluster analysis, and dividing the circumferential joint region into several regions along the horizontal direction; identifying the upper and lower edges of the circumferential joint in each region, and calculating the center point as the center point of the circumferential joint in that region; sequentially connecting each point and extending to the image edge to form a broken line segment, thereby achieving segmented fitting. This invention achieves high-precision, adaptive segmented fitting of non-linear circumferential joints in shield tunnels, improving the accuracy and robustness of circumferential joint detection.

We provide a dashcam with a wide recording range and few blind spots, a display device for the dashcam, and a program. [Solution] The drive recorder 10 includes a main body case 13 that houses a control unit and recording means, and is equipped with a first camera 26 and a second camera 27 that capture an area of ​​more than a hemisphere. The two cameras are mounted in the housing 25 so as to capture opposite directions. When the main body case 13 that houses the control unit, etc., is attached and fixed to the windshield, the vehicle's dashboard and pillars are captured at the edges of the images captured by the two cameras.

The application provides an image noise estimation method and system based on image edge characteristics. The image noise estimation method comprises: obtaining an original image, dividing the original image into a plurality of sub-images to be estimated; judging each sub-image to be estimated based on a gradient image of the sub-image to be estimated to determine whether the sub-image to be estimated is a valid sub-image to be estimated; performing convolution operation or correlation operation on the valid sub-image to be estimated by using a second-order difference filter kernel to obtain a second-order difference image; and estimating noise in the original image based on the second-order difference image to obtain a noise estimation value of the original image. The method obtains a sub-image to be estimated containing more noise and uses the valid sub-image to be estimated to estimate the noise of the original image, thereby accurately estimating the distribution of the noise in the original image. The system can realize the above method.

The embodiment of the present application provides a kind of image edge extraction method and device, it is related to image processing technical field, the method comprises: obtaining X-ray image;Edge feature extraction is carried out to the X-ray image, and initial edge image is obtained;For each pixel point in the initial edge image, the gradient direction corresponding to the pixel point is obtained according to the pixel value of the pixel point and the pixel value of the pixel point in the first preset neighborhood, the pixel value of the pixel point is adjusted according to the obtained gradient direction, and intermediate edge image is obtained;Based on the first edge intensity threshold and the second edge intensity threshold of preset, the pixel value of each pixel point in the intermediate edge image is adjusted, and the edge image of the X-ray image is obtained.The image edge extraction scheme provided by the embodiment of the present application can improve the accuracy of the obtained edge image.

The present application belongs to the technical field of endoscope and image processing, and in particular to an endoscope image anti-shake method based on motion vector compensation, comprising the following steps: collecting a sequence of continuous video frames, performing stability screening on the initial feature point set, eliminating unstable feature points located in the edge region, high light reflection region and low contrast region of the image to form a reliable feature point set, performing robust estimation processing on the original motion vector set based on the matched feature points in the reliable feature point set to generate a stabilized output frame, wherein the remapping adopts a bilinear interpolation algorithm; by constructing a closed-loop anti-shake system taking local feature motion vectors as input, spatio-temporal domain joint filtering as core and sub-pixel remapping as output, high-precision separation and compensation of high-frequency micro-shaking in endoscope imaging are realized without introducing additional mechanical structures.

This invention relates to the field of image recognition, specifically to an automated tunnel construction monitoring system based on image data. The system includes an edge computing node that acquires the original tunnel image and constructs an atmospheric scattering physical model using measured data from an optical dust concentration sensor. The edge computing node transforms the original image to the frequency domain, extracts low-frequency brightness features and high-frequency texture features using discrete cosine transform, and adaptively compensates the high-frequency texture features with weights based on the transmittance matrix calculated by the physical model to suppress diffuse light interference. The compensated features are then input into a lightweight convolutional neural network via inverse transform. This scheme introduces dust physical parameters as prior constraints into the frequency domain feature weight calculation, reducing the probability of image edge artifacts and color distortion under conditions of localized strong light and high dust, ensuring the physical authenticity of target edge features in the reconstructed image, and improving the accuracy of feature extraction under harsh conditions.

The application provides a RAA residual error compensation method suitable for a curve trajectory SAR, comprising: using a radius angle interpolation algorithm to perform imaging processing on obtained SAR echo data, using a new residual error modeling method to perform error modeling, and designing a residual error compensation method to perform phase error compensation to obtain a final SAR image. The radius angle interpolation algorithm is introduced into airborne curve trajectory large forward-looking SAR imaging, and under the condition of large forward-looking, there is a large approximation in the derivation process of the algorithm, there is a large residual phase error for scene edge point targets, and the imaging quality of image edge points is poor. The application compensates the residual phase error of the initial imaging result by constructing a residual phase error space model, improves the focusing depth of the algorithm under the condition of the curve trajectory large forward-looking, expands the effective imaging scene, and greatly improves the limitation of the residual error on the large forward-looking and large scene imaging.

This invention discloses a method, apparatus, device, and medium for controlling ink volume at the edges of printed images. The method involves acquiring the image to be printed and its description; inputting the description information into a pre-constructed image parameter database for matching to obtain image edge information parameters; dividing the image to be printed into regions based on these parameters to obtain at least one image region; and extracting pixels from each edge region using proportional values ​​to obtain a first processed image; further, extracting pixels at intervals from the outermost layer of pixels corresponding to the image to be printed based on the edge information parameters to obtain a second processed image; and finally, determining the ink volume control image for the printed image. This method solves the problems of uneven edges and small peaks in printed images, achieving precise control of ink volume at the edges of printed images, further improving the flatness of the edge regions, and enhancing printing quality.

This application discloses a method, apparatus, electronic device, and readable storage medium for determining a three-dimensional surface shape, belonging to the field of image processing technology. The method includes: filtering a gradient image of a real object image to obtain a filtered gradient image; smoothing the filtered gradient image to obtain a smoothed gradient image; determining a residual image between the gradient image and the smoothed gradient image; gaining the residual image and superimposing the gained residual image onto the filtered gradient image to obtain a gained gradient image; and determining the three-dimensional surface shape of the real object image based on the gained gradient image. This achieves noise reduction while repairing image edges and details, and the three-dimensional surface shape of the real object image can be accurately determined based on the repaired gained gradient image.

The embodiment of the present disclosure provides an image processing method, device, server and storage medium, the method comprises: acquiring a first image and a second image in a virtual reality (VR) image, one of the first image and the second image is a left eye image, and the other is a right eye image; determining a disparity map between the first image and the second image, and determining an edge image of the second image according to the disparity map and the second image, wherein the edge image is an image of the second image except for an image coinciding with the first image; determining to-be-transmitted data according to the first image, the edge image and the disparity map, and sending the to-be-transmitted data to a display device, so that the display device performs restoration processing on the to-be-transmitted data, obtains a restored first image and a restored second image, and displays the restored first image and the restored second image, which can greatly reduce the data amount and further reduce the requirement for an encoder.

The application relates to a kind of electrical measuring instrument detection methods based on visual recognition, comprising the following steps: detecting image edge point in pre-processing image, calculating the center coordinates of image edge point, identifying the corner point coordinates of image edge point using center coordinates, determine the instrument scale distribution in pre-processing image using corner point coordinates;Detect the character mark in pre-processing image, identify the starting scale in instrument scale distribution;Detect the center of the instrument in pre-processing image, extract the pixel point area in pre-processing image across the center of the instrument, obtain the target pixel point area;Determine the short end of the pointer in the target pixel point area using the center of the instrument, determine the long end of the pointer in the target pixel point area using the short end of the pointer and the center of the instrument;Determine the scale measurement value of the long end of the pointer based on the instrument scale distribution and the starting scale, and take the scale measurement value as the detection result of the electrical measuring instrument.The application can improve the detection accuracy of electrical measuring instrument.

The invention belongs to the technical field of computer vision, and particularly relates to an end-to-end shape recognition method and system based on a complex network. The method comprises the following steps: carrying out dense discretization on the edge of a to-be-identified image to obtain an edge description point set; selecting a series of key points on the obtained edge description point set at equal intervals; extracting local features of a key point surrounding edge description point set by using a shape context; taking each key point as a node, defining edges in the network according to spatial correlation among the key points, and constructing a multi-layer complex network by adopting a dynamic evolution strategy; local features and topological features of the complex network are processed through a graph convolutional network, and robust expression of the shape classification task is obtained; and performing classification through a full connection layer network to obtain a classification result, and realizing end-to-end shape recognition. According to the method, the adaptability and robustness of the method to different shape classification tasks are enhanced.

The application relates to the technical field of optical imaging, and discloses an optical lens splicing imaging method and system, the system comprising a calibration module, a preprocessing module, a registration module, a correction module and a fusion module; the registration module, the correction module and the fusion module are arranged, a mixed strategy of global mutual information guidance and local robust matching is used, registration error problems caused by sparse feature points in low-texture areas are solved, the registration precision is effectively improved, a global model based on a Zernike polynomial is used to compensate optical aberration of each sub-lens, the image quality of an edge area of a spliced image is effectively improved, a multi-resolution pyramid optimization mode is used to reduce the calculation complexity of global registration, real-time imaging requirements are met, and a splicing seam is effectively eliminated, and the overall smoothness of an image is significantly improved.

This invention belongs to the field of image segmentation technology and discloses a method, system, medium, device, and terminal for extracting identification documents. The system utilizes an embedded device to acquire multispectral images of the identification document; it models the document image edge segments based on their straight-line geometric properties and infers from the contextual information of the global image; it extracts global image features using a ResNet network and then encodes them using a Deformable DETR encoder; in the first-stage decoding process, it predicts the document edge segments using an attention mechanism and learnable line segment and position queries; in the second-stage decoding process, it compares the correlation between line segment features and image features to predict the relative order between line segments, and obtains the complete edges, vertices, and bounding boxes of the document image through perspective transformation. This invention uses a two-part matching method to predict line segments, avoiding pre- and post-processing, simplifying the detection channel, and achieving true end-to-end processing.

The application discloses an arc length control method, system, device and medium for electric arc additive manufacturing, and relates to the field of additive manufacturing; the control method comprises the following steps: acquiring image data of a target part; determining arc length data by using an image edge data detection method according to the image data; obtaining a fuzzy error by using fuzzy mapping according to the arc length data and set reference arc length data; determining a control quantity according to the fuzzy error; the control quantity comprises a pulse load ratio; and the deviation of the arc length data of the target part in the manufacturing process is compensated according to the control quantity, and the target part is manufactured by using a non-gas shielded welding method; and the application can realize automatic control of the arc length.

The application discloses a kind of image edge detection methods of biological retina multiscale receptive field iteration in the field of image processing and computer vision, including obtaining image to be handled, and pre-processing image to obtain input image.The application is by with center point belonging to an edge edge point in local slice and center point is classified as first edge, the rest is classified as second edge;Subsequently, with the tangent of center point, local slice is divided, and in the partition, from center point outwardly emit ray, if the direction of ray exists second edge point, then the average value of image value in the region within second edge is assigned to the region outside second edge, form compensation slice, the compensation operation is corrected to the pixel distribution in the detection process in adjacent second edge direction, so that the response of center point when subsequent second derivative is reconvolved is weakened, the positioning stability under adjacent interference is improved, positioning deviation caused by adjacent edge interference, multiscale result inconsistency and other problems are avoided.

This invention discloses an event-visual-inertial semantic simultaneous localization and map building method, which is applicable to navigation and localization of mobile robots, unmanned vehicles and drones. This method acquires event data streams from event cameras and inertial data from IMUs, and optionally standard camera images. Based on the event data, it constructs an event activity surface, performs coarse-to-fine event corner detection on it, constructs an event representation for event optical flow estimation, and estimates the event optical flow. It then uses the event optical flow to perform temporal tracking of event corners to obtain multi-time-stack related observations. Based on the event corners and event representations, it extracts descriptive information, performs loop closure detection and relocalization, and introduces loop closure constraints as additional residual terms into a sliding window graph optimization. Within the sliding window, it combines residuals from events, images, IMUs, edge detection, and loop closure relocalization, and sets adaptive weights for multi-source fusion to obtain a six-degree-of-freedom pose sequence. It performs target detection on the event data stream, outputting detection boxes, categories, and confidence scores. Based on the event optical flow, it implements missed detection compensation and motion consistency checks to identify dynamic target regions, eliminates dynamic corners and their observation constraints, and generates an object-level semantic map based on pose estimation. This improves the robustness of localization mapping and environmental understanding in complex lighting, high-speed motion, and dynamic interference scenarios.

This application provides a silicon wafer pose determination method, apparatus, and silicon wafer pose adjustment device, relating to the field of semiconductor or photovoltaic material processing, and solves the technical problem of how to accurately correct the pose of a solar cell. The silicon wafer pose determination method includes: acquiring multiple corner region images of a silicon wafer region in a backlight image; determining the edge lines of each corner region image; determining first position information of multiple first corner points of each corner region image based on the edge lines; determining second position information of corresponding points of each first corner point in the backlight image based on the first position information; acquiring the position information of the image edge line and a preset center point of the backlight image; and determining the rotation angle and displacement of the silicon wafer based on the second position information, the image edge line, and the preset center point. This method can accurately calculate the rotation angle and displacement of the silicon wafer, thereby precisely adjusting the silicon wafer pose.

The application discloses a panoramic image determination method, device and equipment and a storage medium. The method comprises the following steps: acquiring initial images collected by multiple cameras, performing edge detection on the initial images to obtain image edges, performing image space alignment and image splicing based on the positional relationship of the initial images to obtain an initial panoramic image; determining an overlapping area in the initial panoramic image based on the image edges of the initial images, performing pixel adjustment on the overlapping area in the initial panoramic image to obtain a standard panoramic image; and performing pixel fusion on the overlapping area in the standard panoramic image based on pixel feature point matching to obtain a target panoramic image, which can solve the problem that the information described by the existing panoramic image determination method is not clear and accurate, and a panoramic image with clearer, more complete and more accurate information description is obtained, thereby improving the experience of users.

The present application relates to the technical field of image processing, and particularly relates to a kind of edge contour identification method and system for deburring of aluminum alloy die casting, the method comprises: collecting multiple surface images of aluminum alloy die casting under different wind directions;Determine multiple groups of edge straight lines in each surface image;In any surface image, according to the gray scale distribution characteristics and gradient distribution characteristics of each pixel window on each group of edge straight lines, determine the suspected burr area;According to the image feature difference of suspected burr area in two surface images in each surface image group, determine the stability parameter of each suspected burr area;According to the stability parameter of each suspected burr area, determine the real burr area, and carry out image segmentation to the real burr area, extract the edge contour of burr.The present application overcomes the inherent defects of traditional single image edge detection method, such as weak response to tiny burr and easy to be disturbed, significantly improves the accuracy and reliability of detecting burr.

This invention relates to the field of automated microbial antimicrobial susceptibility testing technology, specifically to an automatic device for measuring the diameter of inhibition zones. The device includes an image acquisition module, an antimicrobial susceptibility test strip center positioning module, a radial transmittance profile construction module, a diffusion-MIC inversion module, and a diameter output module. This invention reconstructs the traditional inhibition zone measurement paradigm based on image edge detection and geometric circle fitting into a radial profile inversion paradigm based on a physical model of antimicrobial substance diffusion in agar. It constructs a radial transmittance profile and a family of azimuth profiles with the center of the antimicrobial susceptibility test strip as the polar coordinate origin, inversely infers the local diffusion coefficient field, and uses Fick's second law to nonlinearly invert the minimum inhibitory concentration (MIC) radius. Simultaneously, it separates dual-zone resistant subgroups through residual multi-peak detection and achieves dynamic prediction through temporal differential constraints. This invention achieves a measurement standard deviation better than 0.1 mm, supports dual-zone identification and culture endpoint prediction, and is significantly superior to existing automatic measurement schemes based on edge detection.

The present application relates to the field of image recognition, and more particularly to a tunnel automatic construction monitoring system based on image data. The system includes an edge computing node that acquires original tunnel images and constructs an atmospheric scattering physical model in combination with measured data from an optical dust concentration sensor. The edge computing node converts the original images to the frequency domain, extracts low-frequency brightness features and high-frequency texture features using discrete cosine transform, and performs adaptive weight compensation on the high-frequency texture features using a transmittance matrix calculated by the physical model to suppress diffuse light interference. The compensated features are input to a lightweight convolutional neural network after inverse transformation. This scheme introduces dust physical parameters as a priori constraint into frequency domain feature weight calculation, reducing the probability of image edge artifacts and color distortion under local strong light and high dust superposition conditions, ensuring the physical authenticity of the reconstructed image target edge features, and improving the accuracy of feature extraction under harsh working conditions.

The present application relates to the technical field of image correction, in particular to a thermal imaging image correction method and system based on feature correlation analysis, comprising the following steps: constructing a thermal imaging time sequence, calculating the temperature difference of adjacent frame pixels by pixel difference for the time sequence to generate a change gradient distribution, judging the continuity of the change gradient distribution and counting the number of continuous consistency to divide a trend interval, calculating and generating a gradient distribution result according to the trend interval division result, constructing a mapping based on the gradient distribution result and generating a correction mapping diagram, performing correction on the surface temperature field and redistributing the pixel position and temperature. In the present application, the number of continuous consistency of temperature change gradient is counted to divide a trend interval and perform frame-by-frame weighted update calculation, which completely gets rid of the dependence on external calibration board and feature point extraction, effectively overcomes the matching error caused by image edge blur, deeply adapts to the dynamic temperature fluctuation and thermal response difference of equipment, and practically enhances the remapping precision and gray correction consistency.

The invention relates to the technical field of image sensors, in particular to a pixel array and an image sensor. The pixel array comprises a plurality of pixel units which are distributed in an array mode, and each pixel unit comprises a source follower. Wherein each pixel unit comprises a middle unit and an edge unit, the middle unit is located in the middle of the pixel array, the edge unit surrounds the middle unit, the source follower corresponding to the middle unit is a middle follower, the source follower corresponding to the edge unit is an edge follower, and the width-to-length ratio of the edge follower is larger than that of the middle follower. The gain effect of the edge follower is greater than that of the middle follower, and the gain effect of the optical signal of the edge follower is higher than that of the middle follower, so that the image signal of the edge follower is gradually enhanced, and the definition and the brightness of the image edge are further improved.

A kind of visual-based door rubber strip dynamic detection method and device, the method includes the following steps: image acquisition step: target image with door rubber strip as target object is obtained as input image;Image processing step: input image is processed, and binary image is obtained;Edge recognition step: binary image is Canny operator edge, and the edge in image is identified;Edge contour information output step: the rubber strip image of the edge extracted is compared with rubber strip template picture in rubber strip image with the edge point template matching mode of the target matched with rubber strip template picture is searched, and the edge contour information including possible defect information is output.Thereby, the dynamic detection of rubber strip can be realized, while not affecting the normal production line, with higher accuracy, the defect detection of target is realized.

The present application relates to a kind of remote sensing image super-resolution reconstruction method based on cascade sparse Mamba, belong to remote sensing image super-resolution reconstruction technical field.The method includes: low-resolution remote sensing image is input into remote sensing image super-resolution reconstruction network, in shallow feature extraction part, the shallow feature of low-resolution remote sensing image is extracted, and shallow feature map is obtained;In first level reconstruction part, shallow feature map is input into first level sparse visual Mamba group, and first level deep feature map is obtained;In second level reconstruction part, first level deep feature map is input into second level sparse visual Mamba group, and second level deep feature map is obtained;Shallow feature map, first level deep feature map and second level deep feature map are fused, and high-resolution reconstruction image is obtained.The present application is aimed at solving the technical problems that the existing technology has blurred edge of reconstruction image, texture loss, poor perception quality and the difficulty in balancing global modeling and computational efficiency.

This invention discloses an image processing circuit for autonomous vehicles, belonging to the field of image processing technology. It solves the problems of unexpected refraction and scattering of light due to physical defects in optical microlenses and color filters, and charge coupling caused by parasitic capacitance in pixel units, leading to leakage of optical / electrical signals from a single pixel to adjacent pixels, resulting in color aliasing and blurred image edges. The circuit includes a main image sensor module, a reference image sensor module, and an analog signal preprocessing and multiplexing module. This invention coordinates the synchronous signal acquisition of the main and reference image sensors through a low-jitter clock distribution module. After integration by the analog signal preprocessing module, a dynamic crosstalk compensation module generates an inverse compensation signal based on the reference signal. A high-precision differential correlation dual sampling module then performs noise suppression and crosstalk elimination on the main signal. Finally, a programmable gain amplifier module outputs the purified differential signal.

Embodiments of the application disclose an attack detection method, an electronic device and a computer readable medium. An embodiment of the method comprises: inputting a to-be-tested face image into an attack detection model; dividing the to-be-tested face image into a plurality of image blocks by the attack detection model, extracting block features of the image blocks, processing the block features of the rows and columns of the image blocks, obtaining row features and column features of the to-be-tested face image, processing the row features and the column features, and obtaining an attack detection result; the attack detection result is used to indicate whether the to-be-tested face image is an attack face image that has a difference in image edges from a real face image. The embodiment can effectively defend against edge difference type face attacks and improve the speed and accuracy of attack detection.