5432 results about "Grayscale" patented technology

The present invention relates to the field of image processing. Provided are a screen defect detection method and apparatus, and a device and a storage medium. On the basis of the embodiments of the present invention, the method comprises: acquiring a grayscale image of a screen to be subjected to detection; on the basis of each pixel and surrounding pixels thereof in the grayscale image, determining a gradient image corresponding to the grayscale image; in a gradient direction of the gradient image, moving a window of a set length at a preset step length, wherein the window comprises an odd number of gradient values, and if an intermediate gradient value in the window is the minimum gradient value in the window, setting as a background pixel the pixel corresponding to the minimum gradient value, so as to obtain a target image; and performing segmentation on the target image, and determining, as a defect area on said screen, a target foreground area obtained by means of segmentation. In the gradient direction of the gradient image, a normal pixel is determined on the basis of the gradient of each pixel, and the normal pixel is set as a background pixel, such that a defective pixel in said screen is highlighted.

The invention relates to the technical field of image data processing, and discloses a few-sample scene-oriented industrial product surface defect image analysis method, which comprises the following steps of: obtaining surface gray level image data of a product to be analyzed, calculating a structure tensor matrix and generating an anisotropy degree graph; searching similar blocks in a preset search neighborhood, and constructing a local texture data matrix; performing singular value decomposition on the local texture data matrix to extract a main subspace; constructing a projection operator and utilizing the projection operator to carry out orthogonal projection reconstruction on the local image block to generate a reconstructed background image block; according to the method, through an orthogonal subspace projection mechanism, good product textures and defect signals are separated, random noise is removed, meanwhile, high-frequency structural features are completely reserved, and the method is high in robustness, high in robustness and high in robustness. And the defect detection precision of a complex texture surface in a few-sample scene is improved.

The invention relates to the field of image processing, in particular to a copper-aluminum dissimilar metal welding quality detection method based on image processing, and the method comprises the steps: obtaining a gray level image of a welding image of copper-aluminum dissimilar metal, constructing gray level co-occurrence matrixes according to a preset direction, and carrying out the clustering of the gray level co-occurrence matrixes in all directions, and obtaining a plurality of clusters; acquiring positions of pixel points in the gray level image corresponding to each cluster to construct a scatter diagram, acquiring a local area according to a preset radius of the pixel points so as to calculate an outlier index and an abnormal possibility of each pixel point in the scatter diagram, marking the pixel points as abnormal points when the abnormal possibility is greater than an abnormal threshold value, acquiring a connected domain formed by all the abnormal points, and performing clustering on the connected domain; and obtaining an abnormal area in the gray level image, and evaluating the copper-aluminum welding quality of the abnormal area in the welding image. According to the method, the abnormal textures in the welding area can be accurately recognized through construction of the gray-level co-occurrence matrix and clustering analysis, and the detection precision of the welding quality is improved.

The present application relates to a real-time online cargo volume recognition method, system, and device, and a medium. The method comprises: collecting, by means of a camera, a depth image and a color image of a set area comprising a cargo; introducing a median filter to remove noise pixels in the depth image, and aligning the color image with the filtered depth image on the basis of internal and external parameters of the camera to obtain an aligned color image; on the basis of color, light, and grayscale differences between the cargo and a non-cargo object, using a pre-trained YOLACT model to perform instance segmentation on the aligned color image, and generating a cargo mask; and performing edge fitting, surface extraction, and background separation of the cargo by means of the aligned color image, the filtered depth image and the cargo mask, so as to determine the volume of the cargo. Thus, the present application has high stability and robustness, reliable data can be obtained regardless of the placement location and the placement direction of a cargo, and the working efficiency can be improved.

The invention relates to the technical field of image data processing, in particular to a cable surface defect detection method and system based on machine vision. The method comprises the following steps: acquiring a surface image of a cable, and converting the surface image into a grayscale image; determining the local complexity of each pixel point; obtaining a plurality of areas of the grayscale image, performing complexity determination, and dividing each area to obtain a plurality of windows of each area; determining a contrast limit threshold value of each window; performing image enhancement by using a CLAHE algorithm to obtain an enhanced grayscale image; and carrying out cable surface defect detection on the enhanced grayscale image by using a defect detection algorithm. According to the method, the CLAHE parameters are adaptively adjusted based on local complexity, the window size and the contrast threshold are dynamically determined in combination with gray and gradient information, discontinuity is corrected and eliminated through boundary similarity, and the quality and reliability of a cable defect detection image are improved.

The invention discloses a fire smoke and fire source accurate positioning method based on image recognition, and the method comprises the following steps: S1, collecting multi-modal data, and carrying out the graying, denoising, histogram equalization and shielding correction preprocessing; s2, inputting the preprocessed data into a vortex sensing network to generate a preliminary detection result of flame and smoke; s3, dividing the scene into dynamic space blocks by using a dynamic space division algorithm and allocating priorities; s4, eliminating non-fire feature interference through an environment feature filter, and generating an optimization candidate region; s5, inputting the candidate area into an event-driven time sequence analysis module, and extracting a key frame to analyze flame and smoke diffusion and time trend; s6, verifying position consistency by adopting a flame and smoke joint detection method, and generating a three-dimensional space positioning result; and S7, generating a fire three-dimensional thermodynamic diagram according to three-dimensional positioning and triggering an alarm mechanism. According to the invention, accurate detection and positioning of fire smoke and a fire source are realized by using a vortex sensing network, an event-driven mechanism and the like.

The invention relates to the technical field of image processing, in particular to an anti-collision beam weld defect detection method and system based on image processing, and the method comprises the steps: carrying out the image collection of a weld region of a produced anti-collision beam, and obtaining a gray image; the method comprises the following steps: performing initial partitioning on a grayscale image, respectively obtaining a local complexity index of each initial sub-block, obtaining at least two adaptive sub-blocks based on the local complexity index of each initial sub-block, and performing adaptive local histogram equalization on each adaptive sub-block to obtain a target grayscale image; the method comprises the steps of performing edge detection on a target grayscale image to obtain at least two edge pixel points, performing frequency domain conversion on the target grayscale image according to a gradient direction of each edge pixel point to obtain a frequency domain image, performing filtering processing and time domain conversion on the frequency domain image to obtain a denoised image, and identifying defects in the denoised image by using a neural network. And the defect detection efficiency is improved by inhibiting the periodic texture in the weld seam image.

The present invention relates to the technical field of oil and gas exploration and development. Provided are a digital core-based shale pore type recognition method and a quantitative characterization method. The digital core-based shale pore type recognition method comprises: on the basis of a gray image of a shale sample to be recognized, recognizing an organic matter and a transition zone to obtain an organic matter recognition result; and on the basis of the organic matter recognition result and the gray image, obtaining an organic pore image and an inorganic pore image as a shale pore type recognition result. The present invention can be used to quantitatively recognize organic pores, inorganic pores and micro-cracks of shale, quantitatively characterize shale pore structures, and analyze structural parameters of the organic pores, the inorganic pores and the micro-cracks, so as to perform statistical analysis of distribution of different types of reservoir spaces of shale, thereby being of great significance for shale reservoir quality evaluation and effective shale oil and gas development method optimization.

The invention relates to the technical field of image recognition, and further relates to a phenotypic character image recognition and extraction method for efficient corn breeding. The method comprises the following steps: step 1, acquiring an image of the same plant canopy in a test field by using an unmanned aerial vehicle carrying a multispectral sensor in a plurality of different growth periods from a corn jointing period to a mature period, so as to obtain a normalized grayscale image; 2, performing gradient operation on a target wave band based on the normalized grayscale image to obtain significant edge intensity, and generating a character sensitive probability field of a corresponding time phase; and 3, presetting an optimal observation period for each target character, selecting a pixel set which has a probability value not less than 0.5 and is judged to be the character type from an image of a time phase corresponding to the optimal observation period, and sequentially measuring the total projection area, the average connected domain area and the average leaf angle to obtain a comprehensive breeding sequence value of the character. According to the method, high-precision probability identification of the effective phenotype area of the grain-leaf is realized.

The invention provides a microscope automatic focusing method and device and a computer readable storage medium, and the method comprises the steps: S1, capturing a laser reflection light spot image on a sample through a camera, so as to determine the focusing position of an objective lens; s2, integrating the light spots through threshold processing and morphological operation, analyzing the gradient of a longitudinal gray distribution map to determine the edge and orientation, and finally identifying the maximum contour and framing the maximum contour with a minimum bounding rectangle to determine the width of the light spots; and S3, according to the linear relation between the light spot width and the defocusing distance and by analyzing the semi-circle orientation and width change of the light spot, controlling the focusing motor to reach the focusing position. The distance and the direction between the objective lens and the focal plane are determined by acquiring the reflected light spot image of the laser irradiated on the sample through the microscope objective lens, and the focusing error caused by the nonlinearity of the spot size when the point-semicircle spot approaches the focal plane is avoided according to the linear relation between the spot width and the defocusing distance. And the focusing process is high in efficiency and precision.

The invention relates to the technical field of image enhancement, in particular to a protector bimetallic strip detection method based on machine vision, and the method comprises the steps: obtaining a surface image of a bimetallic strip, and carrying out the preprocessing of the surface image, and obtaining a gray image; performing threshold segmentation on the grayscale image to obtain at least one segmentation region; obtaining an evaluation coefficient of each segmented region, obtaining an adaptive cutting parameter when contrast-limited adaptive histogram equalization is carried out on each segmented region according to the evaluation coefficient of each segmented region, and carrying out image enhancement on each segmented region in the grayscale image according to the adaptive cutting parameter to obtain an enhanced grayscale image; the corrosion detection result in the bimetallic strip is obtained by using the enhanced gray level image, so that the enhanced image can reflect more effective information, and the surface corrosion detection precision of the bimetallic strip according to the enhanced image is improved.

The invention relates to the technical field of visual navigation, in particular to a robot navigation method based on vision, which comprises the following steps: acquiring a grayscale image to extract dynamic point locations, screening boundary features to reconstruct a feature set, calculating gradient change to generate candidate guide points, converting coordinates to construct a path track set, and outputting a navigation planning path. According to the method, rapid elimination of static backgrounds is realized through difference threshold selection, dynamic region extraction accuracy is effectively improved, combined screening of boundary line feature point frequency and average feature density is combined, key feature stability is enhanced, and the identification capability of image local structure mutation is enhanced by using a variance change rate of a gradient amplitude sequence. Robustness of path guide point extraction is improved through judgment of an abnormal section, dynamic construction of a path track is realized by means of continuous time integration of image coordinates, path identification precision and spatial positioning continuity are optimized, and continuity, precision and real-time performance of navigation path planning are improved.

The invention relates to the technical field of industrial machine vision detection and image processing, in particular to an automobile part production mold surface smoothness detection system based on image enhancement, which comprises a data acquisition module used for acquiring an original grayscale image of the surface of an automobile part mold; performing low-pass filtering processing on the original grayscale image to eliminate imaging thermal noise; the manifold reconstruction module is used for constructing a structure tensor field; reversely deducing a pseudo-curvature field of the mold surface; the adaptive enhancement module is used for generating a corrected image; constructing a texture orthotropic diffusion model; generating a texture reconstruction reference image; the surface metering module is used for calculating the difference between the corrected image and the texture reconstruction reference image and generating a defect saliency image; calculating the surface roughness value of the mold surface; according to the method, the problem that design textures and abnormal scratches are difficult to distinguish in the prior art is effectively solved, and the technical bottleneck that micro defects are easily missed in a complex geometric structure in traditional visual detection is overcome.

The invention relates to the field of image processing, in particular to an on-line monitoring method for shaft forging polishing, which comprises the following steps: acquiring a grayscale image of a shaft forging polishing image, performing edge detection on the grayscale image, extracting an edge connected domain, and calculating edge features of the edge connected domain; clustering the edge features to obtain a plurality of clustering clusters, calculating the abnormal degree of the clustering clusters, and calculating the abnormal performance of each edge connected domain in the clustering clusters according to the abnormal degree; and according to the abnormal performance and the edge features, the suspected defect degree of the edge connected domain of the clustering cluster is calculated, and the polishing degree of the shaft forging is judged based on the suspected defect degree. According to the method, through statistical analysis of the edge features and abnormal performance in clustering clusters, the recognition accuracy of the surface defects is improved.

The invention relates to the technical field of image data processing, in particular to a forge piece surface defect detection method and system based on machine vision. The method comprises the steps of obtaining a gray image of a to-be-detected forging surface; determining a boundary significance weight; determining a path consistency weight; screening the boundary significance weight and the path consistency weight to obtain a final weight; and carrying out local adaptive threshold segmentation on the final feature map to obtain a binary image, and carrying out defect identification on the surface of the forge piece to be detected based on a connected region in the binary image. According to the method, the boundary significance weight and the path consistency weight are constructed and are respectively used for accurately positioning defect edges and verifying structure continuity, texture interference is effectively inhibited, and false alarms are reduced; through adaptive Gabor filtering, the problem of a response blind area of a traditional method is solved, finally two weights are fused to modulate filtering response, and the accuracy of forging surface defect detection is improved.

The invention relates to the technical field of obstacle detection, and discloses an unmanned subway obstacle intelligent detection system based on a multi-mode AI sensor, and the system comprises the multi-mode AI sensor, a preprocessing unit, a data processing unit, and a communication unit. According to the method, in data preprocessing, point cloud filtering denoising, down-sampling burden reduction, image graying contour highlighting and Gaussian filtering denoising all improve the processing efficiency and the detection precision. Efficient data fusion establishes a multi-coordinate system conversion relation, blind area misjudgment is avoided, and the complex environment adaptability is enhanced. A comprehensive obstacle detection strategy covers laser radar multi-link processing and Canny algorithm application of visual images, mutual verification is carried out, and the reliability is improved. When obstacles are detected, real-time display and graded alarm can be achieved, early warning and scheduling control can be achieved by means of network transmission information, braking or avoiding can be achieved according to conditions, operation safety is guaranteed, and the system is reasonable in structure, has adaptability and expandability and assists technology development.

The invention relates to the technical field of image enhancement, in particular to a method for detecting an intestinal lesion area of a child based on machine vision. Carrying out threshold segmentation on the grayscale image to obtain a reference point; clustering the reference points and partitioning the grayscale image according to the shape of a cluster; acquiring edge points and a to-be-optimized area according to the gradient of the pixel points in the block area; obtaining a target point according to the gray difference characteristics of the pixel points and the edge points of the to-be-optimized region; obtaining an abnormal point according to the gray difference characteristics of the target point and other target points in the neighborhood and the gray difference characteristics of the neighborhood of the edge points; and obtaining a gray adjustment range value according to the gray difference characteristics of the neighborhoods of the abnormal points and the edge points in the to-be-optimized region. According to the method, a final gray value is obtained according to a gray adjustment range value and a gray discrete feature of an edge point neighborhood; histogram equalization is carried out according to the final gray value, an intestinal tract detection enhanced image is obtained, and the image enhancement accuracy is improved.

The invention relates to the technical field of image processing, in particular to a high-voltage power supply element defect identification method based on computer vision. The method comprises the steps of collecting a surface image of a high-voltage power supply element; the method comprises the following steps: acquiring a grayscale image and a connected domain thereof, acquiring a main direction of the connected domain, constructing a width sequence based on a length perpendicular to the main direction, acquiring a width change coefficient according to the width sequence, and acquiring ambiguity of a pixel point based on a pixel point neighborhood grayscale value; constructing a window for the edge pixel points, obtaining the correlation of the edge pixel points according to the classification of the connected domains in the window, obtaining the similarity between the connected domains according to the Fourier descriptors of the connected domains, and obtaining the scratch integrating degree of the edge pixel points based on the similarity; and screening the edge points of the scratch area according to the scratch integrating degree of the edge pixel points and the ambiguity of the edge pixel points, and completing defect identification. According to the invention, the surface defect detection efficiency of the high-voltage power supply element is improved.

The invention discloses a radar echo extrapolation method and system based on frequency domain enhancement, and the method mainly comprises the following steps: obtaining and preprocessing a historical radar echo grayscale image sequence, generating a sequence sample through a sliding window, and dividing the sequence sample into a training set, a verification set and a test set; the method comprises the following steps: constructing a frequency domain enhanced U-Net network comprising an encoder-decoder structure, introducing a multi-scale deep convolution structure into an encoder and a decoder, and enhancing frequency domain features by using a frequency domain dynamic attention mechanism in jump connection; inputting the training set into the model for training by adopting a composite loss function comprising intensity weighted loss, frequency domain consistency loss and structural similarity loss; and inputting the test set into the trained model, and outputting a radar echo prediction result at a future moment. The method can be effectively applied to the fields of short-term and temporary weather forecast, severe convection monitoring and the like, and provides more accurate and reliable radar echo prediction support for meteorological disaster early warning.

The invention relates to an intelligent detection method for appearance flaws of an OLED display optical film, and the method comprises the steps: carrying out the gray value difference according to an optical film gray image, obtaining a product region, carrying out the Gaussian filtering processing, obtaining an initial preprocessing gray image, and calculating an overall gray mean value; and performing linear scaling processing on the initial pre-processed grayscale image according to the overall grayscale mean value to obtain a first enhanced image, calculating a maximum grayscale value by using a specific structure element according to the first enhanced image, and taking the maximum grayscale value as a current pixel grayscale value to obtain a second enhanced image, and extracting all defect areas in the product area according to the first enhanced image and the second enhanced image, cutting the original gray level image according to the centroid coordinates of the defect areas, sending the cut defect cut image into a classification network model for category judgment, and outputting bubbles, foreign matters and concave-convex defects. According to the OLED display optical film defect detection device, the OLED display optical film defect detection efficiency and the detection result accuracy and reliability are greatly improved.

The invention discloses a malicious traffic classification and identification method, system and device based on a graph convolutional neural network, and a medium. The method comprises the steps of obtaining a data packet, cleaning data, extracting statistical traffic features of the data, and forming a stream sequence; performing detection and pre-classification on the flow sequence; converting the detected and pre-classified flow sequence into a flow grayscale image; inputting the detected and pre-classified traffic grayscale image into an image convolutional neural network for training, and constructing a malicious traffic classification and recognition model for deep classification and recognition; and adopting the trained malicious traffic classification and identification model to carry out deep classification and identification on the network traffic. According to the method and the device, the malicious traffic conforming to the characteristics can be more accurately screened out from a large number of traffic packets mixed with the malicious traffic.

The invention relates to the field of data processing, in particular to a part machining detection method and system, and the method comprises the steps: obtaining images of a to-be-detected part and a template part on a production line, and carrying out the preprocessing of the images, and obtaining corresponding gray images; using an edge detection algorithm to obtain a gradient magnitude and a gradient direction of each pixel point in the grayscale image, performing non-maximum suppression in a neighborhood of each pixel point, screening candidate feature points, and based on the candidate feature points and a plurality of attributes in the neighborhood, respectively obtaining significance feature values of the candidate feature points of the to-be-detected part and the template part; matching is carried out according to the saliency characteristic values of the to-be-detected part and the template part, the comprehensive matching degree is calculated, whether the image of the to-be-detected part is abnormal or not is judged based on the comprehensive matching degree, and part machining detection is completed; according to the method, the real feature points and the false feature points are effectively distinguished by introducing multi-dimensional features such as significance feature values and shape complexity, and the mismatching rate caused by similar texture structures is reduced.

The invention relates to the technical field of image processing, in particular to a bridge crack calibration method and device based on computer vision, and the method comprises the steps: collecting a grayscale image of a bridge, and extracting all edge contours in the grayscale image; obtaining each suspected edge contour; extracting a skeleton line of each edge contour, and determining a first evaluation value of each suspected edge contour; the irregularity and the edge collaboration degree of each suspected edge contour are calculated, and a second evaluation value of each suspected edge contour is determined; and determining a discrimination coefficient of each suspected edge contour, evaluating each suspected edge contour, and calibrating the crack of the bridge. According to the invention, the detection error of other interferences on the bridge crack can be reduced, the accuracy of detecting the crack on the bridge is improved, and the bridge crack is calibrated more accurately, so that the safe use of the bridge is ensured.

The invention discloses a cross-modal pedestrian re-identification method based on a multi-scale cross attention Transform, and the method comprises the following steps: step 1, image acquisition, step 2, feature set acquisition, step 3, feature fusion, and step 4, feature determination: in an initial stage of a network, two standard data sets are used to acquire RGB and infrared images, and then, the RGB and infrared images are subjected to feature fusion; the method comprises the following steps: extracting key features from RGB and an infrared image by using MSCAViT to obtain two feature sets, then converting the RGB image into a grayscale image, inputting the grayscale image into an MSCAViT model to obtain a third feature set, determining an optimal weight by using an OSCSO method through weighted fusion of the three feature sets, finally obtaining a fused feature, inputting a result feature into a decoding end of the MSCAViT, and finally obtaining a final image through the decoding end of the MSCAViT. According to the method, features finally used for image matching are generated, the model is evaluated through multiple indexes, compared with a traditional method, and the result shows that the proposed method shows a more excellent effect on solving the modal difference problem in a VI Re-ID task.

A method for a quality detection of tunnel lining through ground-penetrating radar based on self-supervised learning. In the method, a grayscale image of the tunnel to be detected is obtained, the grayscale image is then input into a trained feature recognition model, to obtain a feature atlas is corresponding to the grayscale image, and then a quality recognition result of the tunnel to be detected is determined according to the feature atlas. In the present application, the feature recognition model is obtained by means of the self-supervised learning according to an unlabeled image set and a labeled image set, and the unlabeled image set is directly utilized to train the recognition model, which not only improves the efficiency of training the tunnel recognition model, but also improves the accuracy of tunnel quality detection results.

The invention discloses a welding quality visual detection method, and relates to the technical field of weld defect detection, and the method comprises the steps: obtaining a pixel neighborhood block corresponding to each pixel point in a to-be-detected welding gray image, and obtaining the illumination influence degree in each pixel neighborhood block according to the gray feature and gradient feature of each pixel point in each pixel neighborhood block; according to the position distribution characteristics between the edge end points of the adjacent edges in the welding gray level image to be detected, the same kind of edges are judged; obtaining a gamma value adjustment coefficient of a pixel point at the central position in each pixel neighborhood block by obtaining the confidence coefficient that each similar edge is the inner edge of the welding seam and the defect degree of each similar edge; the gamma value of each pixel point is adjusted, the gray level output value of each pixel point in each pixel neighborhood block is obtained, and an optimized welding gray level image is obtained; performing quality detection on the to-be-optimized welding grayscale image; the accuracy of welding quality detection can be improved.

The invention relates to a single cell image segmentation method based on a minimum circumcircle, which solves the technical problem of how to improve the precision, robustness and adaptability of an image-based single cell segmentation method, and comprises the following steps of: firstly, obtaining an original cell image, and secondly, converting the original cell image into a gray level image; preprocessing and binarization processing are carried out on the grayscale image to obtain a binarized image, a cell edge contour image is obtained through processing, a minimum circumcircle is drawn for each contour in the cell edge contour image to obtain a cell circumcircle image, and finally, the center of the minimum circumcircle is used as the center point of a cutting area to obtain a cell edge contour image. And determining the boundary of the cutting area by taking the background including the cells and within a certain range around the cells as a standard, cutting the cell circumcircle image, and finally obtaining a single cell image. The method is suitable for single cell identification and segmentation in a microscope image, and can be widely applied to the fields of cell biology research, medical diagnosis, drug screening and the like.

The invention relates to the field of image processing, in particular to a goat milk filtering efficiency detection method and system based on machine vision, and the method comprises the steps: obtaining a gray-scale image of the work of a filter screen, carrying out the image enhancement of the gray-scale image, and obtaining a preprocessed filtering image; calculating the blocking degree of the filtered image; and when the blocking degree is larger than or equal to a preset threshold value, the filter screen is cleaned. By combining the stability and smoothness of a flowing path and the change of gray values at different moments, when the blockage degree reaches or exceeds a preset threshold value, the system can automatically trigger the cleaning process, so that the continuous smoothness of the filter screen and the filtering efficiency of goat milk are ensured.

The invention discloses a smart mine safety production violation behavior identification method, system and device and a medium, belongs to the technical field of smart mine safety identification, and aims to solve the technical problem of how to improve the accuracy and efficiency of mine safety production violation behavior identification, realize real-time and accurate safety supervision of the whole process of mine operation and improve the safety of mine safety production violation behaviors. According to the technical scheme, the method comprises the following steps: data acquisition and preprocessing: installing a camera in a key operation area of a mine to acquire video image data, and carrying out denoising, graying and normalization preprocessing operation on the video image data to obtain preprocessed video image data; and constructing a deep learning model based on a convolutional neural network: introducing an attention module into the network structure of the deep learning model, and training the deep learning model by using the marked video image data including the safety production violation behavior and the normal operation behavior, and adopting a transfer learning method in the training process.

The invention relates to the technical field of electronic component detection, in particular to an IC (integrated circuit) carrier plate detection method based on surface state image extraction, which comprises the following steps: acquiring a gray image, analyzing structural parameters, extracting gradient features, detecting boundary disturbance, integrating the image, calculating an abnormal score, identifying a defect position area, extracting features and outputting an identification result. According to the invention, by analyzing the structure parameters of the bonding pad in the gray level image, calculating the edge line segment, the center coordinate and the spacing, and constructing the two-dimensional coordinate system, the regional positioning reference is enabled to have geometric consistency, the coordinate mapping is combined with the gradient direction change frequency and the continuous aggregation point, the boundary disturbance identification precision is improved, and the image division is executed based on the disturbance region. According to the method, non-functional region mixing is effectively avoided, a clustering and probability model is introduced after region gray level statistics, a deviation scoring mechanism is constructed, gray level feature abnormity is accurately recognized, the discrimination capability of small-amplitude and low-contrast defects is improved, and the selectivity and target focusing performance of feature detection are enhanced.