1050 results about "Texture feature" patented technology

The disclosure discloses a surface defect detection method, system, equipment, and terminal, comprising: S1, LBP operator: before comparing the benchmark image with the actual shot image, a LBP texture feature extraction algorithm is first used for pre-processing; S2, Sift feature-point matching: adding a Sift feature-point matching algorithm to calculate key feature points in the image and compare; S3, defect detection: when the local feature points of the shot image are successfully matching with a certain image in the benchmark image library, a detailed comparison will be made and complete defect detection. The disclosure integrates various image processing methods in advance without any setting by users. After running the system with one click, it can automatically perform detection, greatly reducing personnel training costs. The disclosure has simple structure and low cost. The basic structure only needs a computer, camera, and light source, making it easy to use and with highly mobile.

The invention relates to an injection molding product defect detection method based on machine vision, which comprises the following steps: collecting material information of a to-be-detected injection molding product in real time, and dynamically matching and adjusting light source parameters according to spectral reflection characteristics of materials to ensure image collection quality; secondly, the collected images are preprocessed, edge features and texture features are extracted, a three-dimensional model is constructed through multi-view image splicing, and three-dimensional defect features are extracted; thirdly, the multi-dimensional features are input into a deep learning model, the defect probability is calculated through feature fusion and forward propagation, and whether the product has defects or not is judged; if the defect exists, further identifying the defect category, and calculating the number and size of the defect; and generating a standardized detection report based on the defect information. According to the method, the image adaptability of products made of different materials is improved through dynamic light source adjustment, the two-dimensional and three-dimensional features are fused, the defect recognition accuracy is improved, and full-process automation from qualitative judgment to quantitative analysis of the defects is achieved.

The invention belongs to the technical field of image analysis, and discloses an old people health state prediction and alarm method based on AI image analysis, and the method comprises the steps: obtaining a depth image, a color image and a pressure image of an activity environment of old people, carrying out the combined analysis of the depth image and the pressure image, recognizing a terrain semantic region in a monitoring region, and carrying out the prediction of the health state of the old people; and carrying out dynamic segmentation on the gait contour of the old people. And the terrain gait coupling degree is calculated by analyzing the depth change characteristics and the motion period characteristics of the gait sub-regions, so that the abnormal gait sub-regions are screened out. And for the abnormal regions, the terrain gait coupling degree, the dynamic texture features of the color image and the pressure distribution features of the pressure image are fused, and a cross-modal health risk vector is constructed. Based on the vector, the health state of the old people is accurately predicted, and a corresponding alarm signal is generated. According to the invention, precise evaluation and timely early warning of the health state of the old people are realized, and intelligent guarantee is provided for the home safety of the old people.

The invention provides a multispectral fusion-based valve optical image enhancement identification method and system in severe weather, and the method comprises the steps: obtaining multispectral optical image data and mechanical dynamic response characteristics of an industrial pipeline valve in severe weather; based on the mechanical dynamic response characteristics, separating the multispectral optical image data to extract multiband texture characteristics associated with the reflection characteristics of the valve material; double amplitude correction of a spatial domain and a frequency domain is carried out on the multiband texture features; distributing a spectrum fusion weight for the multispectral optical image data according to the corrected multiband texture features; and generating an enhanced optical image matched with the actual mechanical state of the valve based on the spectrum fusion weight and the corrected multiband texture features. According to the method, the recognition accuracy of defects such as valve surface cracks and deformation in severe weather is improved.

The invention relates to the technical field of strategy analysis, and discloses an alpine grassland rapid restoration strategy analysis method, system and device and a medium. The method comprises the following steps: performing quality evaluation and restoration on an alpine grassland multispectral image to obtain a standard image set; spectral texture features are extracted, degradation classification is carried out, and a feature set and a score table are obtained; evaluating the degradation degree in combination with multi-dimensional indexes, and generating a distribution matrix and an index set; forming a repair scheme library and a parameter list through case analysis and parameter optimization; analyzing a restoration effect in combination with environmental data to obtain an index matrix and an evaluation table; and optimizing the monitoring data to obtain a strategy scheme and an adjustment parameter set. According to the method and the device, the degradation conditions of alpine grassland of different types and different areas are accurately identified, the restoration strategy can be dynamically adjusted according to the real-time monitoring data, and the restoration effect is improved.

The invention discloses a black pig image segmentation method based on multi-feature fusion, and relates to the technical field of image processing, and the method comprises the steps: obtaining a black pig image, and carrying out the contrast enhancement processing; inputting the enhanced image into a residual convolutional network to generate a depth feature map, extracting a local shape feature map through curvature threshold segmentation and a double fitting strategy, and fusing the two feature maps to generate a black pig feature map; establishing a spatial position prior probability graph based on the black pig sample library, calculating regional correlation and performing adaptive weighting to obtain a fusion feature graph; boundary segmentation and iterative optimization are carried out on the fused feature map based on the dynamic behavior pattern map and the attitude constraint rule, and an initial segmentation map is generated; and adopting a group behavior model as an optimization criterion, correcting the boundary of the initial segmentation image, and outputting a final segmentation result. According to the method, the segmented enhancement function based on the double-peak characteristic and the local texture feature self-adaptive adjustment strategy are constructed, so that differential enhancement of image preprocessing is realized.

The invention relates to the technical field of digital oil painting optimization, in particular to a multi-level visual texture feature optimization method and system for a digital oil painting. The method comprises the following steps: acquiring an original digital oil painting image, performing regional image semantic deep mining, performing illusion perception noise abstract injection, and constructing an illusion perception texture feature layer; performing local continuity texture visual detection on the illusion perception texture feature map layer, performing stroke tension transmission analysis, and constructing a stroke tension perception texture map; performing multi-scale semantic region division and inter-region cross shielding mining on the stroke tension sensing texture map, and constructing a multi-layer texture detail expression map; and carrying out layer-by-layer texture time sequence drawing speculation on the multi-layer texture detail expression map, and carrying out time sequence texture visual rendering optimization to generate a time sequence layered texture rendering effect. According to the method, through multi-layer texture reconstruction, oil painting textures of different time sequence levels are improved, and the stereoscopic visual effect is embodied.

The invention relates to the technical field of computer vision, in particular to a defect segmentation positioning method and system for an inorganic mineral casting image, and the method comprises the following steps: calling an illumination image to analyze brightness, matching exposure parameters, splicing the image, analyzing a gradient, recognizing a defect, screening an effective region, calculating a gray variance, and constructing roughness weight recognition texture features. According to the method, the high-reflection area identification, the brightness gradient analysis, the pixel-level roughness weight and the structure tensor analysis are combined, the exposure interval can be dynamically adjusted when the casting image is processed, the defect type information is output in the direction, and the positioning information is generated by correcting the recognition position in combination with the actual coordinate of the target spot. The method has the advantages that the high-reflection area identification, the brightness gradient analysis, the pixel-level roughness weight and the structure tensor analysis are combined; the method has the advantages that the method is simple and easy to implement, detail loss of overexposure areas is reduced, the recognition precision of defect areas is improved, accurate area segmentation and classification processing are achieved, roughness weight calculation combining gray variance and pixel density is combined, the sensitivity to surface fine defects is enhanced, and the precision and reliability of defect positioning are improved.

The invention relates to the technical field of image detection, and discloses a printing defect detection method and system based on computer vision. According to the method, a multi-mode imaging device collects a printing surface multi-channel fusion image, distortion is corrected through a dynamic segmentation module, a detection area mask is extracted, and a standard image is output. Inputting the standard image into a flaw detection model, extracting microscopic lattice points and macroscopic texture features, and matching with a template library to generate a probability distribution diagram; and performing multi-level judgment according to the image, quantitatively evaluating the flaws, and finally outputting flaw coordinates with space-time labels and type labels. The system comprises a multi-source acquisition module, a geometric correction module, a feature matching module, a grading decision-making module and a closed-loop control module, and real-time detection and feedback control of printing flaws are achieved. The method improves the detection precision and efficiency, is suitable for different types of printed matters, and has a wide application prospect.

The invention discloses a laser powder bed forming defect identification method based on mask segmentation optimization. The method comprises the following steps: S1, acquiring image data of a forming area in a laser powder bed forming process in real time by using imaging equipment; s2, obtaining preprocessed image data; s3, extracting key features in the image based on the preprocessed image data by adopting a multi-scale feature extraction method; s4, generating a corresponding defect area mask; s5, performing feature extraction on the defect area according to the generated defect area mask; and S6, classifying and evaluating the extracted defect region features, determining the type and severity of the defect, and forming a defect identification result. According to the method, the texture features are calculated through the gray-level co-occurrence matrix, the edge features are extracted through Canny edge detection, and the defect types are comprehensively judged in combination with regional morphological analysis, so that the recognition accuracy of different defects is improved.

The invention relates to the technical field of image enhancement, in particular to an archive restoration system based on image enhancement, and the system comprises a high-frequency detail enhancement module which receives a damaged archive image, processes high-frequency information through a high-pass filter, and obtains a high-frequency processed image; the high frequency processed image is sharpened while maintaining smoothness of the low frequency region, generating a detail enhanced image. According to the method, the high-frequency information of the damaged archive image is independently processed, the detail part in the image is independently extracted and sharpened, and meanwhile, the low-frequency region is kept smooth, so that the detail definition of the image and the overall visual naturalness are synchronously improved; and in the image texture repairing process, the texture of the undamaged area is analyzed, and key texture features are extracted and expanded, so that the texture and the pattern of the damaged area are accurately reconstructed, and the integrity and the authenticity of the image structure and the content are ensured.

The invention relates to the technical field of three-dimensional image reconstruction, in particular to a VR-based three-dimensional reconstruction method and system for textile culture heritage, and the method comprises the steps: obtaining a multi-view image collection sequence of the textile culture heritage, carrying out the geometric topological feature extraction of the multi-view image collection sequence, generating a three-dimensional point cloud geometric model, and carrying out the reconstruction of the three-dimensional point cloud geometric model. Performing multi-resolution texture feature mapping processing on the three-dimensional point cloud geometric model to generate an initial three-dimensional model fusing geometric topological features and multi-resolution texture features; performing dynamic geometric optimization processing on the initial three-dimensional model based on preset VR display parameters to generate a target three-dimensional model after surface continuity correction; and performing dynamic matching processing on the target three-dimensional model and a preset VR interaction algorithm to generate a VR visual model with a multi-dimensional interaction attribute, and transmitting the VR visual model to a VR display terminal to perform three-dimensional space rendering, so that the three-dimensional reconstruction and interaction intelligence degree of the textile culture heritage can be improved in combination with a VR technology.

The invention provides an aluminum profile defect analysis method and system based on texture features, and relates to the technical field of edge detection, and the method comprises the steps: obtaining an aluminum profile surface image, and determining an extrusion direction; dividing the image into a plurality of local blocks, carrying out gradient direction and amplitude calculation on each block, and judging whether the block belongs to a high-confidence-coefficient texture region or not by combining a direction difference value and confidence coefficient; a first suppression coefficient is executed on the high-confidence-coefficient texture region for primary reduction to form a first processing image, then the updated image is judged again, a second suppression coefficient with higher strength is applied to the region still having obvious texture features, and a second processing image is generated; and finally, defect identification is carried out on the weakened image through edge detection, and residual texture false edges are removed in combination with direction consistency or connectivity analysis. The method has the advantages of light weight, low computing power consumption and high accuracy, and can be applied to online detection and quality control of the surface flaws of the aluminum profile in industrial production.

The invention discloses an image multi-modal feature extraction method, a ground feature classification and recognition method and a GIS image generation method. Comprising the following steps: firstly, extracting texture features from a target image by adopting a multi-directional statistical method fusing rotation invariant coding of a local binary pattern and a gray-level co-occurrence matrix, and extracting color features from the target image by adopting an LAB-HSV dual-color space collaborative analysis method to obtain features of different modes of the target image; and then according to the information values of the extracted color features and texture features, adjusting the weight ratio corresponding to the color features and the texture features so as to optimize the recognition precision of the classification model on complex ground features. And finally, according to a weight ratio corresponding to the color feature and the texture feature, performing weighted fusion on the color feature and the texture feature to obtain a corresponding multi-modal feature vector.

The invention relates to the technical field of data visualization, in particular to a multi-modal fusion anorectal data visualization analysis method and system, and the method comprises the following steps: synchronously collecting an anorectal image and a pressure signal, aligning a timestamp, carrying out the preprocessing, outputting a standardized image and a coded physiological signal, and extracting edge and texture features. After LSTM coding, splicing with image features, and fusing into a joint feature map; generating a multi-scale heat map by the feature pyramid network, and coarsely positioning a lesion; after boundary optimization, U-Net recovers details, and a pixel-level lesion probability graph is output; superposing the Jet color gradation to the original image in a semitransparent manner, and marking a pressure abnormal time period; and performing weighted scoring to generate a fourth-level clinical suggestion. According to the anorectal disease diagnosis method, through multi-modal space-time alignment, multi-scale feature fusion, boundary optimization and a quantitative scoring system, the problems of data splitting, insufficient precision and low efficiency of a traditional method are solved, a high-precision and high-robustness intelligent diagnosis tool is provided for anorectal diseases, and the scientificity and efficiency of clinical decision making are remarkably improved.

The invention provides a rail turnout foreign matter detection and type analysis method and system, and the method comprises the steps: arranging a first camera group at the front end of a train, recognizing the position of a turnout through a preset target detection model, and calibrating foreign matters in the turnout through a preset foreign matter detection model; shooting an image calibrated to the foreign matter in the turnout through a second camera group as a to-be-analyzed image; extracting a second boundary region rolled by a wheel on the foreign matter from the to-be-analyzed image through a preset wheel rolling track detection model, performing coarse classification on the foreign matter in the to-be-analyzed image to obtain a coarse classification result, and selecting a boundary detection strategy to identify the first boundary region according to the coarse classification result in the foreign matter image; and respectively extracting corresponding texture features in the first boundary region and the second boundary region through a deep learning network, performing one-to-one mapping on the extracted texture features to obtain a mapping relation group, performing analysis through the mapping relation group to obtain a feature change rate, and performing analysis through a multi-feature analysis strategy to obtain a specific foreign matter type.

The invention belongs to the technical field of artificial intelligence image generation, and particularly relates to an image generation method and system based on style feature injection, and the method comprises the steps: carrying out the multi-dimensional feature analysis of a reference image specified by a user side through a pre-trained multi-level style extraction model, extracting a style feature vector, and carrying out the feature extraction of the style feature vector; the style feature vector comprises a color feature vector, a texture feature vector, a composition feature vector and an illumination feature vector; obtaining a text description input by a user side, and outputting triple information including a scene entity, an entity attribute and a spatial relationship; performing weighted fusion on the triple information and the style feature vector to obtain a style enhanced semantic embedding vector; and inputting the style-enhanced semantic embedding vector into a generative adversarial network for splicing, and outputting an image. The method has the effect of remarkably improving the similarity between the generated image and the reference style.

The invention relates to the technical field of computer vision, in particular to a station platform limit intelligent detection method and system based on computer vision, and the method comprises the following steps: obtaining platform region spectral data, extracting spectral features according to the reflectivity value of each pixel point in a spectral range, comparing the spectral features with the features of various known materials, and determining the position of the station platform limit. And identifying material types of the plurality of positions, detecting obstacles, and generating material classification feature values. According to the method, the spectral data is acquired, the reflectivity value is compared to identify the material type, the texture feature and the gradient change rate are utilized to analyze and identify the fuzzy area, misjudgment on the fuzzy identification is reduced, point cloud data and image information are adopted for time-space synchronous registration, and multi-source sensing information is fused to enhance the time-space consistency of three-dimensional reconstruction. The size accuracy of the platform model is improved, the space offset is calculated by comparing engineering design drawing data, a structure deviation quantitative evaluation system is established, a deformation area is positioned, and the foreseeability of safety early warning is improved.

The invention discloses a stone cultural relic disease identification method based on a hyperspectral image, and aims to solve the problem that stone cultural relic disease investigation depends on artificial experience and is time-consuming and labor-consuming. The method comprises the steps that a hyperspectral image of a stone cultural relic to be explored is collected and preprocessed; extracting spectral features; selecting a central wave band of three wave bands of blue light, green light and red light to construct a true color image; extracting texture features and color features of the true color image; fusing the spectral features, the texture features and the color features to obtain combined features; establishing a typical disease feature database; training a disease classification model by using a random forest, a support vector machine, a gradient elevator and a convolutional neural network; carrying out typical disease identification by using the intelligent identification model and drawing a disease cloud picture; and calculating the typical disease area proportion of the to-be-detected area to carry out disease evaluation. According to the invention, in-situ, lossless and rapid identification of stone cultural relic diseases is realized, manpower and material resources are saved, and an intelligent identification method is provided for cultural relic protection exploration work.

The invention relates to the technical field of image enhancement, in particular to a field-level crop image enhancement system based on multi-source remote sensing data, and the system comprises an artifact joint detection module which analyzes the mutation or periodicity of a statistical mean value and a variance value in the row and column directions of an input field-level remote sensing image, and recognizes an artifact stripe region. According to the invention, the method achieves the high-precision removal of thin clouds and stripe artifacts through carrying out the mutability analysis of the local mean and variance of the input remote sensing image in the row and column directions, positioning a stripe artifact region, and building an attenuation numerical relationship in combination with the spectral information of an adjacent clear sky reference region, and reduces the ground feature information loss in the image. Multi-scale coefficient decomposition is further performed on the image after artifact removal, texture features are extracted from homogeneous crop canopies in adjacent undamaged areas, and high-frequency coefficients of damaged areas are dynamically replaced and reconstructed, so that the texture structures of the damaged areas are highly consistent with those of the adjacent areas, and the regional texture continuity and authenticity are improved.

The invention discloses a visual defect detection method for precise and complex parts, and particularly relates to the technical field of part detection. The method comprises the following steps: dividing the surface of a precise complex part into a plurality of to-be-detected areas, calculating a gray level co-occurrence matrix characteristic value, texture characteristic distribution and a spectrum analysis result by combining image data under multi-angle and multi-light-source conditions, further comparing surface bulges and gray level difference characteristics of a normal area and the to-be-detected areas, and calculating the surface defect similarity. On the basis of surface texture complexity and defect similarity, risk misjudgment is performed through fuzzy logic analysis, a detection area is divided into high, medium and low risk areas, a detection algorithm is optimized to improve accuracy, finally, algorithm performance is verified through test data, a parameter adjustment range is limited when the misjudgment rate is increased, systematic deviation and performance reduction are avoided, and the detection accuracy is improved. Therefore, the detection stability and the overall quality control reliability are improved.

The invention belongs to the technical field of hydraulic engineering detection, and particularly provides a hydraulic metal structure surface coating defect detection method and a hydraulic metal structure surface coating defect detection device based on image processing. In a wet and dry alternating environment of a hydraulic metal structure, a multispectral imaging device is used for acquiring an image of a metal structure surface coating, and the acquired multispectral image is preprocessed; a complete and clear image to be detected is obtained; extracting features related to coating defects from the preprocessed to-be-detected image, wherein the features comprise spectral features, texture features and color features; and inputting the extracted features related to the coating defects into a pre-trained classification model, and classifying the coating defects through the classification model. According to the method and the device, the state of the coating can be comprehensively described, and the type and the degree of the coating defect can be accurately identified.

An image processing method, performed by a computer device, comprising: extracting sketch texture features at multiple scales from a sketch image; extracting image noise features at multiple scales from preset noise; determining color guide information corresponding to the sketch image; encoding, for each scale, a noise feature based on a sketch texture feature and the color guide information to obtain multi-scale image features; and performing multi-scale decoding on these image features to obtain a colored image comprising a sketch texture corresponding to the sketch image and a color based on the color guide information. A related training method and apparatus are also provided to develop models for this image processing technique.

The invention relates to the technical field of image segmentation, in particular to an automatic segmentation method and system for an echocardiogram of the department of cardiology, and the method comprises the following steps: based on image data of the echocardiogram, extracting gray level distribution, edge feature and texture feature information, analyzing gray level change amplitude, screening gray level change abnormal regions, and recognizing connectivity features. According to the method, the segmentation accuracy is effectively improved by extracting image gray, edge and texture features and identifying abnormal regions, noise and artifact interference are reduced by optimizing low connectivity regions, and the segmentation accuracy is improved. A problem area is analyzed and positioned in combination with multi-frame gray level change, a segmentation result is adjusted, the processing stability and consistency are enhanced, meanwhile, an optimized alarm node is output based on a frequency trend, more accurate and stable heart image analysis is supported, and the clinical application practicability is improved.

The invention discloses a lightweight rice leaf disease identification method based on YOLO target detection. The method comprises the following steps: S1, inputting a rice leaf disease image; s2, carrying out super-resolution reconstruction on the image by using an improved Real-ESRGAN algorithm; s3, inputting the picture into an improved YOLOv8n target detection algorithm; s4, obtaining disease spot category and position information; and S5, visualizing the information on the image. The invention relates to the technical field of target detection, and has the beneficial effects that image super-resolution reconstruction is carried out aiming at the problems that a rice leaf scab target is relatively small and an image acquired in real time is relatively fuzzy, so that the resolution of the small target is improved, and the definition and texture features are improved. On the basis of a super-division model Real-ESRGAN, a group of residual dense modules containing five layers of cavity convolution layers are designed to help the network to acquire receptive fields and information of different scales.

The invention provides an appearance defect detection method and device and application thereof, and belongs to the technical field of intelligent detection, and the method comprises the steps: carrying out the preliminary detection of a to-be-detected product based on the texture features of an initial detection image; performing background and foreground separation processing on the initial detection image to obtain an effective detection image; determining a defect doubt area in the effective detection image; and based on the gray values of all the pixel points in the defect doubt area, judging whether a gray abnormal defect exists or not so as to realize final detection of appearance defects of the to-be-detected product. According to the method, a three-section composite detection process of texture abnormity pre-detection, background separation and defect post-processing verification is carried out on the detection image, so that the problem of inaccurate detection caused by the structure and material characteristics of a product and environmental interference is effectively solved, and the accuracy and robustness of defect detection are remarkably improved.

The invention provides a high-noise-resistance bearing fault diagnosis method for complex working conditions, relates to the technical field of bearing fault detection, and constructs a progressive technical system of'dynamic noise reduction-multi-modal fusion-cross-domain association '. According to the method, firstly, working condition self-adaptive noise reduction is achieved through sliding window dynamic principal component analysis, and redundant noise caused by rotating speed fluctuation is stripped while periodic impact characteristics are reserved; meanwhile, on this basis, a time domain-frequency domain multi-mode complementary feature system is constructed, and the denoised signals are respectively input into a bidirectional gating circulation unit and an improved residual network added with FFT transform to extract bidirectional time sequence dependence and frequency domain texture features. And finally, performing feature enhancement on a time sequence signal and a time-frequency image extracted by the multi-modal feature network by adopting eight self-attention, constructing a dynamic interaction channel by taking a time sequence feature as Query and an image feature as Key-Value, and adaptively screening a cross-modal association feature most related to the context of the current rotating speed.

The invention belongs to the technical field of image processing, and discloses a nonferrous metal target detection method and system for a scraped car. According to the method, through deep coupling of a GAN bimodal fusion network and a double-backbone network, a scraped car nonferrous metal target detection model is constructed, so that nonferrous metals in scraped cars are efficiently and accurately sorted. A bimodal fusion network is introduced into a model input layer, a fusion image with infrared thermal saliency and visible light texture features is generated, and the problem of cross-modal information splitting is solved; according to the dual-backbone network, standard convolution is replaced by a three-branch structure of an MGHCM module, large target contours, middle target semantics and small target details are synchronously captured, and self-adaptive shunting and efficient fusion of multi-scale features are achieved in cooperation with dynamic feature routing of a DHFBlock module; and the detection head network is optimized into a rotating frame prediction and cross-modal cross entropy classification mechanism so as to adapt to the placement scene of the metal fragments at any angle.

The invention discloses a quality detection system based on three-dimensional point cloud scanning and RGB image fusion, relates to the technical field of industrial automatic detection, and aims to solve the problem of high quality detection error rate caused by the fact that a detection result is easily interfered by visual angle change and illumination conditions in an existing detection method. According to the method, the color point cloud model under the unified coordinate system is constructed through space calibration and point pixel level registration, and the structural information and texture features of the workpiece are effectively reserved. Therefore, the problem that the quality detection error rate is high due to the fact that the detected object has curvature dramatic change, a strong reflection area or frequent shielding is solved. According to the method, the detection accuracy is improved, meanwhile, the robustness and the operation efficiency are high, the industrial workpiece detection requirements with complex curved surface structures and high surface quality requirements can be met, and the actual application requirements of an industrial site for an intelligent surface detection system are met.

The invention relates to the technical field of aluminum alloy surface detection, and discloses an aluminum alloy surface defect recognition method and system based on image recognition, and the method comprises the steps: carrying out the image collection of an aluminum alloy surface, and obtaining an original image of the aluminum alloy surface; preprocessing the collected original image to obtain a preprocessed image; performing feature extraction on the preprocessed image, and extracting geometric features, texture features and color features of defects; screening the extracted features by adopting a feature selection method based on the combination of a genetic algorithm and a support vector machine to obtain a feature subset; inputting the feature subset into a GNN-GAN model, outputting the type, position and size of a defect according to a defect identification result, and marking and recording the defect; according to the invention, the labor cost and the time cost required by manual detection are reduced, the product rework and the rejection rate caused by missing detection are reduced, and the quality stability of the aluminum alloy product is ensured.