237 results about "Lab color space" patented technology

The invention discloses a two-stage network underwater image enhancement method based on color correction and multicolor space stretching, and the method comprises the steps: compensating a red channel and a blue channel through a green channel, obtaining a red channel correction image and a blue channel correction image, constructing an improved U-Net network, and outputting the weight mapping of the red channel and the blue channel; performing dot product multiplication on the red and blue channel correction images and the corresponding red and blue channel weight mapping, and splicing the dynamic compensation results of the red and blue channels and the green channel into an RGB space color correction image; the RGB space color correction image is converted into an HSI space and a Lab color space respectively, histogram stretching is carried out on the HSI space, the Lab color space and the RGB space respectively, finally, the three kinds of images subjected to histogram stretching are converted into the RGB color space respectively, and the three images converted into the RGB color space are spliced to obtain a comprehensive input image; and inputting the comprehensive input image into the improved U-Net network, training again, and outputting an enhanced image, so that the imaging resolution and the color correction effect of the underwater camera can be remarkably improved.

The invention discloses a product surface paint layer color difference detection method and system. The method comprises the following steps: converting an original image from an RGB color space to an LAB color space; then foreground segmentation is carried out to realize extraction of a to-be-detected object; lAB values of foreground pixels in the LAB image are extracted, a K-means algorithm is used for clustering the LAB values of the foreground pixels, and color grading of the foreground pixels is achieved; using a region merging denoising function to optimize the clustering labels, eliminating noise points with too small areas, and enabling the noise points to be automatically fused into regions with peripheral color levels; chromatic aberration between the clustering centers is calculated, and clustering center pairs with the distance larger than a threshold value serve as significant chromatic aberration pairs; and covering masks of different colors in each stage of color area, and generating a visual result of color grading and a visual result of an area with remarkable color difference. According to the invention, accurate and efficient image color difference detection is realized, and the method has a wide application prospect.

The invention provides a three-dimensional setting printing correction method and device, equipment and a storage medium, and the method comprises the steps: scanning a three-dimensional setting, and obtaining a three-dimensional point cloud model, color grid data and a multispectral texture mapping graph; based on the three-dimensional point cloud model and the color grid data, determining an offset defect labeling graph and a parallax offset matrix; performing two-dimensional mapping and global correlation calculation on the multispectral texture mapping graph to generate a texture anomaly thermodynamic graph and an LAB color space deviation vector; inputting the parallax offset matrix, the offset defect labeling diagram, the texture anomaly thermodynamic diagram and the LAB color space deviation vector into a preset decision tree model, and generating a root cause classification label and a printing quality score; and based on the root cause classification label and a preset historical printing parameter database, constructing a reinforcement learning model, taking the printing quality score as a reward function of the reinforcement learning model, and outputting a printing correction scheme.

The invention provides an auxiliary pathological analysis method for a microscope system, and belongs to the technical field of pathological analysis, and the method comprises the steps: firstly carrying out the gray normalization, denoising and other preprocessing of an obtained pathological section image, and carrying out the illumination equalization processing through a region adaptive histogram equalization algorithm; then, color features of the image are extracted in an LAB color space, a color representation center is calculated, a feature mapping matrix is established, and meanwhile texture features are extracted and fused with the color features; and carrying out dimension reduction on the fused feature vector by using a principal component analysis algorithm, and reserving important feature information. And then, establishing a convolutional neural network model, inputting the feature vectors after dimension reduction, performing training and cross validation by using a labeling data set, and improving the model performance through iterative optimization. According to the invention, the technical problem that the diagnosis and judgment of a doctor are seriously influenced due to inconsistent color expression caused by the difference of shooting conditions of pathological section images observed by a microscope is solved.

The invention relates to the technical field of image processing, in particular to a printing image-text defect detection method based on machine vision, and the method comprises the steps: obtaining an RGB image of a to-be-detected printed matter, and carrying out the Lab space conversion, and obtaining a target image; performing normalization processing on the L channel value of each pixel point in the target image to obtain a normalized brightness value so as to divide the pixel points in the target image into a preset number of illumination level intervals; obtaining a color fluctuation statistical parameter of any one illumination level interval, and obtaining an optimal color difference threshold value corresponding to any one illumination level interval; according to the optimal color difference threshold value corresponding to the illumination level interval to which each pixel point in the target image belongs, performing Lab color space template comparison on the target image to obtain an image-text color difference defect result of the to-be-detected printed matter, thereby improving the accuracy of image-text printing defect detection based on the color difference threshold values.

The invention relates to the technical field of underwater image enhancement, and provides a deepwater image enhancement method based on multi-color space coupling, which comprises the following steps: determining an optimal channel, a moderate attenuation channel and a severe attenuation channel in an RGB color space based on a pixel average value, carrying out asymmetric histogram cutting pre-correction on the optimal channel, and carrying out deep-water image enhancement on the optimal channel; designing a loss function to enable an attenuation channel to approach an optimal channel through iterative compensation; in the Lab color space, clustering and segmenting the brightness component L into a plurality of pixel blocks by using an unsupervised pixel clustering model PCM, and performing local contrast enhancement and guided filtering-based noise reduction processing on each pixel block; and finally, carrying out nonlinear stretching on the chrominance components a and b by adopting a self-adaptive exponential function to improve the color naturalness. Through a multi-color space cooperative enhancement mechanism, the limitation of single color space processing in a traditional method is effectively solved, and the color correction precision, brightness enhancement and noise reduction balance and the naturalness optimization effect of the deep water image are improved.

The invention relates to the technical field of machine vision and optical detection, in particular to a tea color automatic evaluation method and system based on machine vision. The method comprises the following steps: constructing a stable and repeatable optical measurement environment and calculating radiation baseline parameters; performing multi-band polarization sequence imaging on the tea sample to obtain original data with spectrum and structural characteristics; separating mirror surface and diffuse reflection components on the surfaces of the tea leaves, and performing spectrum consistency correction by combining a radiation baseline; performing multiband spectrum reconstruction and normalization processing on the basis of the corrected diffuse reflection image, mapping to a CIE Lab color space, extracting pixel-level brightness, chromaticity and color distribution statistical indexes, and generating a color feature vector in combination with spatial weighting matrix optimization; a final score is calculated by constructing a quantitative evaluation and grade judgment model and integrating perception color separation, drying calibration color separation and a penalty term, and a grade result is output. According to the method, objective, accurate and automatic evaluation of the color of the tea leaves is realized.

The invention relates to a micro-expression recognition method based on double-flow feature fusion, and belongs to the field of computer vision. The method comprises the steps of obtaining N frames of images of a micro-expression video from a start frame to a vertex frame, and calculating an optical flow field between adjacent frames by adopting an optical flow algorithm so as to effectively extract optical flow features of micro-expressions; the method comprises the following steps: acquiring N frames of images of a micro-expression video from a start frame to a vertex frame, converting the N frames of RGB face images into a CIE Lab color space, calculating a pixel difference between two frames of converted Lab images, and extracting pixel stream features of micro-expressions; and inputting the optical flow features and the pixel flow features into a constructed double-flow three-dimensional convolutional network for feature extraction and fusion, and classifying micro expressions. According to the method, the dynamic and subtle changes of the micro-expression are effectively captured by combining the optical flow and the pixel difference characteristics. By integrating spatial and temporal information, richer feature representations are provided. The improved attention mechanism further focuses on fine facial changes, and the accuracy of micro-expression classification is improved.

The invention provides a textile surface defect detection method and system based on image features, and relates to the technical field of textile surface defect detection.The method comprises the steps that an RGB image of a to-be-detected textile is obtained and zoomed to 512 * 512 pixels, and then at least five semantic tags are output through a semantic segmentation model; the method comprises the following steps: adopting hyperspectral imaging with the wavelength of 400-1700 nm and the resolution of 10 nm, calculating the pixel reflectivity through a reflectivity formula, comparing the reflectivity of 5-8 characteristic wavebands, marking as a pollutant if the average deviation exceeds a threshold value, matching the type and position of a stain with a spectral angular distance, and after a standard sample and a semantic region are converted into an LAB color space, extracting a minimum rotation enclosing rectangle and dividing a grid, invalid pixels are filled with 3 * 3 neighborhood effective pixel weighted mean values, and a grid LAB mean value is calculated. A grid pollutant pixel corresponding to an image to be detected is filled with a standard sample in a mean value mode, then channels A and B are clustered through a sliding window, a pilling area is preliminarily judged according to the standard deviation of a channel L, and the pilling area is confirmed by combining the structured light three-dimensional point cloud and comparing with a threshold value.

The invention discloses a head reconstruction method based on a single photo, and the method employs a high-low model mesh fusion algorithm to guarantee the facial details, achieves the model lightweight, and guarantees that the model can adapt to a mobile device and meet the real-time rendering demands. Through combination of key point detection and a PnP algorithm, accurate projection and extraction of facial textures are realized, and the fidelity of texture details is effectively improved. Besides, a skin color matching method based on a Lab color space is introduced, the problem of chromatic aberration between the face area and the prefabricated chartlet is solved, natural edge transition and vivid visual effect are ensured, and therefore a high-quality head three-dimensional model meeting the requirements of multiple platforms and multiple scenes is generated.

Provided in the embodiments of the present description are an image processing method and apparatus, and a device and a storage medium. The method comprises: acquiring an image to be matted which has a pure-color background; determining a ternary graph corresponding to the image to be matted; selecting a target image from an A-channel image and a B-channel image, in a LAB color space, of the image to be matted, wherein the color of the target image is closer to the background color of the image to be matted; by taking the target image as a guide image, performing guided filtering processing on the ternary graph to obtain an Alpha image; and on the basis of the Alpha image, performing matting to obtain a foreground area from the image to be matted. By means of the solution, details of an edge part of a foreground area can be retained to the greatest extent; in addition, the processing speed of a matting process can be increased.

The invention relates to the technical field of textile detection, and discloses a textile color migration detection system and method. According to the system and the method, textile image data are acquired through a high-precision scanner, an image processing module sequentially performs noise removal, brightness adjustment and contrast enhancement operation on the data, and a color space conversion module converts an image from an RGB color space to a Lab color space based on a contrast-enhanced textile image. The color feature extraction module extracts color features of the textile image, the color features comprise Euclidean distances between a color histogram and colors, calculates a standard value of differences between the colors and transmits the standard value to the migration detection module, and the migration detection module sets a color migration threshold value; and the color migration detection module is used for comparing with a standard value of color difference to judge whether the color of the textile is migrated or not, generating a color migration detection report and transmitting the color migration detection report to the visualization module, and the visualization module displays the color migration detection report on a visualization terminal.

The invention provides a facial skin flaw enhancement method based on a Lab color space. The method comprises the following steps: firstly, acquiring an RGB face image and converting the RGB face image into a CIE Lab color space with uniform perception; then, performing differentiation treatment according to the manually selected skin flaw type: for the vascular flaw, extracting statistical characteristics of a component and driving adaptive nonlinear transformation, and generating a grey-scale map which highlights the red flaw; for pigment flaws, nonlinear transformation is carried out on the component L, and then collaborative linear weighting and feature amplification are carried out on the component L, the component a and the component b, so that a grey-scale map with highlighted pigment spots is generated. And finally, coloring the grey-scale map in the Lab color space through adjustable parameters to generate a high-contrast color enhanced image. The method overcomes the dependence on hardware and training data in the prior art, can clearly and adaptively enhance various flaws such as acnes, couperose streaks and color spots, shows robustness under different illumination, and can be widely applied to clinical beauty, later photography and real-time video processing.

The invention belongs to the field of computer vision, and particularly relates to a cloud cover density calculation method combining image features and superpixel partition merging and a corresponding device thereof. The method comprises the following steps: 1, preprocessing an original image P0 containing the sky and the cloud layer to obtain a gray level image H1, an HSV image P1, a Lab image P2, a color enhancement image E1 and a gray level enhancement image E2 which correspond to different color spaces; 2, P0 is divided into a plurality of superpixel areas, and background elimination is carried out by combining characteristic information of H1, P1, P2, E1 and E2; gradually combining the superpixel regions belonging to the cloud layer or the sky; and 3, extracting values of an S channel and a Lab color space three channel of the superpixel region, generating a four-dimensional feature vector, and inputting the four-dimensional feature vector into a K-means algorithm for clustering. And 4, calculating the cloud cover density according to the clustering result of the K-means algorithm. The problem that the sky and cloud layer image segmentation task is difficult to process by the existing traditional image segmentation technology is solved.

The present invention discloses a jaundice detection method and system based on visual recognition, and the detection method comprises: obtaining first image data of a person to be detected under white light shooting illumination, and obtaining second image data of the person to be detected under blue light shooting illumination; performing image color extraction on the first image data based on HSV color space to obtain color data of the first image data; performing comparison on the color data of the first image data with the image chromaticity data of a standard image to obtain first comparison data; extracting the proportion of green component and yellow component from the second image data based on Lab color space recognition, and obtaining second comparison data based on shooting parameter settings; obtaining an image yellow chromaticity change evaluation value of the person to be detected based on the first comparison data and the second comparison data, and outputting a detection result based on the evaluation value. By performing image analysis on the yellow component change value, the image detection accuracy is improved, thereby improving the evaluation reliability of the occurrence of jaundice symptoms.

The invention discloses a method for dynamically monitoring the drying quality of fruits and vegetables by fusing chromatic aberration-shrinkage rate analysis. The method comprises the following steps: placing a fruit and vegetable sample on a rotatable objective table, drying according to a preset program, and carrying out multi-angle shooting by using a digital camera according to a timestamp to obtain an RGB color image; the method comprises the following steps: preprocessing by an image analysis system, converting an image into a Lab chromaticity space, calculating a chromatic aberration value delta E and a comprehensive shrinkage rate S, integrating data to generate a change curve, and constructing an appearance form model. In actual drying, real-time data and the model are compared, and early warning is carried out when the color difference value exceeds 3.5 or the shrinkage rate deviation exceeds 15% and image acquisition is abnormal. The drying equipment is provided with a high-precision electric rotary objective table, a specific annular light source array, a high-resolution camera and a standard calibration color card, and illumination and color measurement accuracy is ensured through calibration. Through multi-step operation, accurate and dynamic monitoring of fruit and vegetable drying quality is achieved, and drying process optimization is assisted.

The invention relates to the technical field of LED control, and discloses a multi-mode LED swimming pool lamp control method, system and device. The method comprises the following steps: calculating switching path pre-calculation data based on a remote controller mode switching instruction and color characteristic parameters of E light modes; according to the switching path pre-calculation data, carrying out progressive modulation on the RGBW four-channel PWM duty ratio to obtain a linear transition modulation sequence; on the basis of the linear transition modulation sequence, trajectory calculation is carried out on dynamic change mode color change in a CIE Lab color space, and color transition trajectory parameters are obtained; and closed-loop control is carried out on RGBW four-channel LED currents according to the color transition track parameters, PWM control signals are obtained, progressive adjustment of the PWM duty ratio is achieved, and the visual abrupt sense and the flicker phenomenon caused by traditional abrupt change type switching are eliminated.

The invention discloses a colorimetric analysis method of ion concentration detection test paper suitable for a mobile phone App, and belongs to the technical field of test paper colorimetric analysis. The method comprises the following steps: obtaining pictures after a test paper reference color block and a reagent block react with a to-be-detected ion solution through a camera, ensuring the quality of the acquired image through user operation guidance, performing color correction on the test paper image, performing complete reference color block ROI region image detection and extraction based on contour shape features, and determining the detection accuracy of the test paper based on the collinearity of the reference color block. The method comprises the following steps: performing reagent block ROI region image positioning extraction according to a full contour and other position geometrical relationship, extracting color information of a test paper reference color block and a reagent block color development region, calculating color similarity in RGB, HSV and Lab color spaces based on the color information, and performing test paper color identification according to a voting mechanism. The method can provide convenience for patients with color vision defects, has certain self-adaption to different light environments, can reduce the influence of subjectivity and the like on the accuracy of detection results, and is simple and rapid in detection process.

The invention discloses a non-contact heart rate estimation method and system based on multi-modal signal fusion, and relates to the technical field of image processing, and the method comprises the steps: positioning a region of interest in a face video, extracting HSV and Lab color space signals, and fusing the HSV and Lab color space signals; suppressing ambient light interference through background brightness compensation to obtain a color conversion signal and normalizing the color conversion signal; adjusting the bandwidth filtering signal of a band-pass filter by using the historical heart rate to generate an rPPG signal; performing time domain and frequency domain analysis on the rPPG signal, and performing weighted averaging by combining two results to obtain a heart rate estimation value; a historical data smoothing strategy is adopted to correct a heart rate estimation result to ensure accuracy. According to the method, high-precision non-contact heart rate estimation is realized through extraction and fusion of HSV and Lab color space signals, background brightness compensation, signal normalization and processing of human face video signals based on a historical heart rate adjustment band-pass filter.

The invention discloses a nozzle voltage adjustment method and device based on camera automatic identification printing color blocks and a storage medium, and relates to the technical field of printing device calibration, and the method comprises the steps: controlling a printing device to output a pattern containing each nozzle independent test area and color blocks; acquiring chromaticity distribution data through an image acquisition device, and extracting characteristic parameters such as brightness and saturation; determining a reference characteristic parameter based on a reference selection rule, and calculating a deviation value between each test area and the reference; an adjustment instruction is generated according to the mapping relation between the deviation value and the voltage adjustment amount, verification printing and iterative optimization are carried out after implementation until the precision is met, and through dynamic selection of a reference color block, LAB color space chromatic aberration calculation, voltage sensitivity coefficient matrix mapping and aging compensation, the adjustment precision and efficiency are improved, the dependence on artificial experience is reduced, and the adjustment precision and efficiency are improved. The method is suitable for nozzle calibration of high-precision printing equipment.

The invention discloses a solder paste state detection method, device and equipment and a readable storage medium. The method comprises the steps that a to-be-detected solder paste area image and a corresponding template solder paste area image are acquired; the to-be-detected solder paste area image and the template solder paste area image are pseudo depth images; the to-be-detected solder paste area image and the template solder paste area image are converted into a Lab color space, and the similarity between the first color texture feature of the converted to-be-detected solder paste area image and the second color texture feature of the converted template solder paste area image is determined; and determining the state of the to-be-detected solder paste corresponding to the to-be-detected solder paste area image according to the similarity. According to the scheme, the solder paste state is determined based on the color change in the pseudo depth image, the height information participates in solder paste state detection on the basis of the two-dimensional image, the accuracy of solder paste state detection is improved, dependence on a depth camera is not needed, and therefore the solder paste state detection cost can be reduced.

The invention discloses a printed cloth defect detection method, and belongs to the technical field of printing. The method comprises the following steps: acquiring an image of to-be-detected printed cloth, and determining the height of a printed loop; intercepting an image containing n patterns from the acquired image as a standard image, wherein n is greater than or equal to 1; converting the standard image into a Lab color space and extracting an L channel to obtain a first image; segmenting the first image to obtain a first image sub-image, and counting and storing the sum of each column of pixels of the first image sub-image; acquiring at least one to-be-detected image of the printed cloth from the acquired images; the height of the to-be-detected image is consistent with that of the standard image; converting the to-be-detected image into a Lab color space and extracting an L channel to obtain a second image; segmenting the second image to obtain a second image sub-image, and counting the sum of each column of pixels of the second image sub-image; and detecting flaws of the printed cloth based on the sum of the columns of pixels of the first image sub-graph and the sum of the columns of pixels of the second image sub-graph.

The invention provides a tomato leaf health status assessment system and method, and relates to the technical field of tomato leaf health assessment, and the method specifically comprises the steps: firstly, carrying out the preprocessing of a tomato leaf image, converting the image into a Lab color space, and carrying out the segmentation and extraction of a leaf region based on a chromaticity threshold value; and then leaf area chromaticity channel statistical magnitude is counted, pixel-level anomaly detection is carried out through an adaptive threshold value, and an abnormal area mask and a normal area mask are generated. Performing connected domain analysis on the abnormal region, extracting local color and texture features, and extracting corresponding features of the normal region as a reference; the health deviation degree is obtained by calculating the feature distance between each abnormal region and the normal region, and the comprehensive health assessment index is generated by combining the area proportion of the abnormal regions after normalization. And finally, judging a health state according to a preset threshold rule and outputting an evaluation result. According to the invention, a comprehensive and accurate decision basis is provided for agricultural producers, and the automation level and evaluation accuracy of tomato disease monitoring are improved.

The invention discloses an eye disease analysis method based on an eye image, and the method comprises the steps: carrying out the image segmentation of a sclera region through employing a convolutional neural network architecture u-net network model of image segmentation, carrying out the recognition of a hyperemia part of the sclera region, and calculating the proportion of the hyperemia part through employing FrangiFilter blood vessel recognition in combination with the color recognition of an LAB color space; by combining the image algorithm with a traditional image processing and deep learning model, the segmentation precision of the sclera region, the identification precision of sclera congestion, the robustness, the high accuracy and the rating accuracy are improved under the condition that the operation speed is not reduced.

The invention discloses an ultrasonic image enhancement method based on an ultrasonic Gaussian kernel and non-sharpening mask sharpening. The method comprises the following steps: acquiring an original image; removing the noise of the original image by using an ultrasonic Gaussian kernel to obtain a denoised image; performing Gaussian blur processing on the denoised image to obtain a Gaussian blur image; extracting high-frequency information in the original image; sharpening the extracted high-frequency information part to obtain a sharpened image; performing weighted fusion on the sharpened image and the original image to obtain a weighted image; the weighted image is converted to a Lab color space, and the brightness channel contrast is enhanced; and restoring the contrast-enhanced image to the RGB color space to generate a final enhanced image. According to the method, the Gaussian kernel denoising and the non-sharpening mask sharpening technology are combined, noise in the ultrasonic image is effectively removed, image details are enhanced, and the identifiability of the image is further improved through self-adaptive contrast enhancement.

The invention discloses an art design intelligent color matching auxiliary system and method based on artificial intelligence, and relates to the field of art design. Comprising an image analysis module, and the image analysis module comprises a noise reduction preprocessing unit for carrying out noise suppression on an input image by adopting a bilateral filter and a semantic segmentation unit for analyzing an image semantic region and outputting a segmentation mask by using a DeepLabv3 + model; a color space conversion unit that inputs and converts the RGB image into a CIELAB space; a style feature extraction unit; a palette optimization module; and a rendering output module. According to the method, the four-dimensional statistical vector of the LAB color space is extracted, high-level features such as the quantized lightness centrality, the chroma discreteness and the distribution skewness are extracted, the problem of feature expression distortion of a traditional method is solved, the style loss function based on the Gram matrix is constructed, gradient descent optimization is combined, and it is ensured that the palette meets the aesthetic rule and style consistency at the same time.

The invention discloses a sandstone slice image component identification method based on a GL-SLIC algorithm and semi-supervised learning. The method comprises the following steps: firstly, acquiring a sandstone slice image amplified by 200 times under single polarization, converting an RGB image into a CIE-Lab color space, and extracting a GLBP texture feature vector; then a GL-SLIC superpixel segmentation algorithm is constructed, and adaptive segmentation based on texture and color features is realized in combination with a Gabor filter and a local binary pattern; then implementing a region merging algorithm to generate complete mineral particles and pore fragments; finally, a classifier based on VGG16 and a discriminator based on ResNet18 are constructed, a semi-supervised self-training framework is adopted, a model is initialized by using about 6% of manual labeling samples, a high-confidence-coefficient pseudo-label extension training data set is generated through iteration, and automatic recognition of sandstone components such as quartz, pores, kaolinite, rock debris and a matrix is achieved. According to the method, the recognition accuracy of 96.3% on a test set is achieved, compared with a traditional method, the segmentation precision and the recognition accuracy are remarkably improved, the data labeling cost is greatly reduced, and an efficient technical scheme is provided for automatic analysis of geological images.

The invention discloses a fabric texture fidelity rendering method and system based on a scanned image and a target color number, and solves the technical problems of poor color perception consistency and texture detail loss. The method comprises the steps of preprocessing a standardized input image through illumination compensation, performing double-space color mapping in an LAB color space, partitioning the image by adopting K-means clustering according to texture features of a gray-level co-occurrence matrix, establishing a color mapping function containing global mapping transformation and local adaptive adjustment, and adaptively determining a fusion coefficient according to regional variance; decomposing the image into a base layer and a detail layer by adopting guide filtering, and performing spectrum constraint on the detail layer to ensure that the spectrum amplitude is consistent with the original; extracting material parameters such as a normal map, a roughness map and a metallity map; according to the method, fabric texture details are completely reserved while color perception consistency is guaranteed, rendering reality is improved, batch processing of multiple groups of target color numbers is supported, and the method is suitable for application scenes such as costume design, e-commerce display and virtual fitting.

The present application discloses an image enhancement method, device, equipment and storage medium under severe weather conditions, and relates to the field of image enhancement. The method reads an image frame and performs LAB color space conversion to determine the degraded image frame and the corresponding image type therein; if it is a dust image, it is converted into a grayscale image and a power operation is performed on the grayscale value to obtain a dust-free image; if it is a haze image, image edge filtering is performed and a transmittance image is extracted; when the filtered image contains a sky area, the transmittance value of the area is extracted and deviation correction is performed to obtain a corrected transmittance image; based on the corrected transmittance image, image defogging is performed to obtain a haze image. The scheme distinguishes the degraded image frames and the corresponding image types caused by severe weather according to the color cast factor and chromaticity, and adopts different image enhancement schemes for different image types, especially for haze images, gradient domain guided filtering and transmittance optimization of the sky area are performed to obtain a clearer defogging image.

The invention relates to the field of computer vision, and discloses an image classification and defect detection method based on self-supervised contrast learning, which comprises the following steps: converting an image to be detected into an LAB color space, and generating feature representation information of principal component analysis based on separated information of a channel A, a channel B and a channel L; extracting a target potential feature vector from the feature representation information through a first encoder; extracting high-level semantic information from the target potential feature vector through a first semantic feature encoder; performing classification detection on the advanced semantic information through a classifier network; wherein the construction of the first encoder and the first semantic feature encoder comprises the following steps: respectively inputting a training sample set into the first encoder and the second encoder; and performing parameter optimization on the first semantic feature encoder and the second semantic feature encoder by adopting a comparative learning mode.