92 results about "Retinex algorithm" patented technology

Provided is a low-illumination image adaptive enhancement method, an apparatus thereof, a device and a storage medium. The method includes: inputting a low-illumination image to be enhanced into a pre-trained low-illumination image adaptive enhancement model to obtain an enhanced image. A method of training the low-illumination image adaptive enhancement model includes: acquiring a training set containing the low-illumination image and the corresponding reference image; constructing the low-illumination image adaptive enhancement model based on a Retinex algorithm. The low-illumination image adaptive enhancement model includes a projection module, an illumination component module, a reflectance component module and an enhancement module. The training set is used to train the low-illumination image adaptive enhancement model, and the trained low-illumination image adaptive enhancement model is obtained. The brightness and the contrast of the image are enhanced, and at the same time, the color and structure information of the image is restored effectively.

The invention relates to the technical field of image recognition data processing, in particular to a pipeline element library creating method based on a model training and image recognition technology, which comprises the following steps of: acquiring pipeline element image data with depth information through a multispectral imaging scheme, executing Retinex algorithm illumination equalization and morphological restoration through a cascade preprocessing pipeline, and establishing a pipeline element library. A standardized data set associated with the metadata is generated. A multi-task joint learning framework is adopted to integrate ResNet-50, HRNet and Mask R-CNN networks, a bottom convolution feature extraction layer is shared, oil stain and strong reflection antagonism data synthesized by a generative adversarial network is injected, model parameters are optimized in combination with a progressive training strategy, and a lightweight MobileNetV3 model is output. Model parameters and an index structure are dynamically updated, and system self-calibration is achieved in combination with a cross-device calibration protocol and data consanguinity tracking. According to the method, through automatic data labeling, multi-task feature multiplexing and retrieval feedback closed loop, the construction efficiency of the pipeline element library is effectively improved, and the recognition error rate in a complex environment is reduced.

The invention relates to the technical field of image processing, in particular to an automatic snow removal operation method and system based on image enhancement processing, and the method comprises the following steps: collecting an original accumulated snow image of a road surface, decomposing an image brightness component into a low-frequency component and a high-frequency component through a multi-scale Retinex algorithm, and generating an enhanced image; on the basis of the enhanced image, calculating a joint gradient of saturation and brightness in a pixel point HSV space, and generating a snow area binary segmentation image by adopting an improved Otsu dual-threshold method; the accumulated snow compactness grade is judged according to the texture uniformity; and driving the execution mechanism to spray the snow removing agent according to the path coordinates, and adjusting the spraying concentration of the snow removing agent in combination with the compactness grade. According to the method, the snow boundary definition and the plaque internal texture identifiability are improved, a high-quality image foundation is laid for subsequent segmentation and analysis, and the adaptive capacity and stability of the system under the natural light condition are enhanced.

The invention discloses a locomotive sanding detection method. The method comprises the following steps: step 1, judging whether a license plate number is identified or not; 2, broadcasting sanding notifications of all the positions according to a preset time difference; step 3, acquiring a video shot by the monitoring camera; 4, dark light enhancement is carried out through a U-Net model, a YOLOV5 model is input, and whether the monitoring camera is shielded or not is judged; step 5, carrying out illumination adaptation processing on a plurality of continuous frames of images of the video obtained in the step 3 through a CLAHE algorithm; step 6, carrying out image enhancement through a Retinex algorithm and a U-Net model; 7, inputting a 3D-CNN model to carry out feature extraction and time sequence modeling; and step 8, inputting a Transform Encoder model, and outputting the probability of each sanding flow rate. According to the method, the detection of the sand spraying amount can be completed, and the detection of whether the monitoring camera is shielded or not can be completed before the detection of the sand spraying amount.

The invention relates to the technical field of image enhancement, in particular to a surface process defect detection method and system for copper foil production, and the method comprises the steps: collecting a copper foil image in a production process; obtaining each image block in the copper foil image; performing edge detection on the copper foil image to obtain each edge line in the copper foil image; determining a texture feature coefficient of each image block, constructing an illumination contrast coefficient of each image block, correcting a standard deviation parameter in a Gaussian filter, and performing image enhancement on each image block by using the Gaussian filter after parameter correction and a local Retinex algorithm; and carrying out surface defect detection on the copper foil through the enhanced copper foil image. Therefore, the precision of copper foil surface process defect detection is improved.

The invention belongs to the technical field of image data processing, and discloses a driver fatigue state early warning method and system based on face recognition tracking, and the method comprises the steps: extracting a face region through semantic segmentation, splitting a single-channel gray-scale map, calculating the local complexity and noise degree of a pixel point, obtaining a window adjustment coefficient, and carrying out the adjustment of the window adjustment coefficient; the size of the adaptive window is determined, different pixel region features can be adapted in a targeted manner, and the processing deviation caused by a fixed window is effectively reduced; an illumination component is subsequently extracted based on an adaptive window, and the image is enhanced through a Retinex algorithm, so that the face image quality under complex illumination can be improved, and the detail definition is improved; and finally, the fatigue degree is detected in combination with a PERCLOS algorithm, so that the facial fatigue characteristics of the driver can be more accurately captured, and effective recognition and early warning of the fatigue state are realized. According to the invention, continuous and stable monitoring can be realized in the driving process, reliable guarantee is provided for driving safety of a driver, and the risk of accidents caused by fatigue driving is reduced.

The invention discloses an autoclaved aerated concrete member surface defect intelligent identification system based on image processing, and particularly relates to the field of defect identification, comprising an image acquisition module, an image preprocessing module, a defect candidate region extraction module, a defect identification and classification module, and a result output and alarm module; according to the method, a high-definition industrial camera is used for collecting a component surface image, and adaptive median filtering and a Retinex algorithm are adopted for image denoising and enhancement, so that the influence of noise and uneven illumination is eliminated; utilizing an improved multi-threshold segmentation and Canny edge detection algorithm to accurately extract a defect candidate region; the method comprises the following steps: extracting three types of feature parameters of shape, texture and gray scale, and inputting the three types of feature parameters into a deep learning model taking ResNet50 as a basic network to realize automatic identification and classification of four types of typical defects of cracks, holes, unfilled corners and surface peeling; and finally, the system divides severity levels according to the defect size, and triggers differentiated visual alarm and linkage control.

The invention relates to the technical field of image processing, in particular to an image data processing method and system for digital management of environmental protection equipment, and the method comprises the steps: collecting a real-time image of an aeration area through a waterproof camera disposed above an aeration tank, detecting the definition of the camera regularly, and carrying out the cleaning when the definition is lower than a preset value; a self-adaptive Retinex algorithm is adopted for the real-time image of the aeration area to eliminate water surface reflection interference, noise caused by water turbidity is reduced through a multi-frame weighted average method, and a processed image is obtained; segmenting a bubble region in the image based on a U-Net network and calculating a bubble distribution parameter; comparing the bubble distribution parameter with a bubble distribution threshold value, and generating an aeration rate optimization instruction according to average dissolved oxygen data obtained by a dissolved oxygen sensor; and controlling the air blower based on the aeration rate optimization instruction. The power of the air blower can be better adjusted according to the bubble condition of the aeration tank, so that more energy is saved.

The invention relates to the field of tea disease detection, in particular to a tea disease detection method based on a lightweight YOLOv8 model. According to the technical scheme, the method comprises the steps of collecting various types of tea disease images; the method comprises the following steps: preprocessing multiple types of collected tea disease images, including illumination normalization, geometric enhancement, illumination enhancement, adaptive histogram equalization, a multi-scale Retinex algorithm, dynamic Gamma correction and image noise removal, and marking scab regions in different types of tea disease images to obtain a training data set; a lightweight YOLOv8 model is constructed; using the training data set to train the lightweight YOLOv8 model; and inputting a collected tea disease image into the trained lightweight YOLOv8 model, and outputting a tea disease type through the trained lightweight YOLOv8 model. The method is suitable for tea disease detection.

The invention discloses a food waste detection method and system based on image processing, belongs to the field of image recognition, and aims to realize efficient and automatic recognition and quantification of kitchen waste. According to the method, residual food images are collected at multiple periods and multiple angles in a kitchen garbage can or a dinner plate recovery area through high-resolution and multi-spectral imaging equipment, preprocessing is carried out in combination with an improved Retinex algorithm and a space self-adaptive denoising technology, and the image quality is improved. Afterwards, fine segmentation of a food area is achieved through a multi-scale super-pixel segmentation and graph segmentation algorithm, and multi-category intelligent recognition is conducted on remaining food through a recognition network fused with multi-modal features. The system further combines stereoscopic vision and Monte Carlo sampling to dynamically and accurately count the volume or weight of various residual foods. The method has the advantages of high adaptability and accurate statistical result, and can provide data support for catering management, resource recovery, nutrition evaluation and the like.

The invention discloses an optical fiber connector defect detection method and system based on deep learning, and belongs to the field of optical communication equipment quality detection. Aiming at the problems of insufficient feature extraction, insufficient utilization of multi-modal data, poor environmental adaptability and the like of a traditional detection method, detection is realized through a multi-modal data fusion-feature enhancement-dynamic calibration closed-loop architecture. An improved Retinex algorithm and the like are adopted for preprocessing data, features are fused through a cross-modal attention mechanism and the like, features are enhanced through YOLOv8, GAN and the like, and a threshold value is dynamically calibrated in combination with Bayesian optimization and the like. Various defects can be recognized, the detection accuracy rate is larger than or equal to 99.5%, the false alarm rate is smaller than or equal to 0.1%, full-process intelligentization of detection is achieved, and the method is suitable for high-precision defect recognition.

The invention discloses a welding defect identification method and device based on artificial intelligence, and relates to the technical field of welding defect identification. Comprising the following steps that S1, continuous frame images are collected to obtain welding appearance images, and accurate time is marked on each frame image; s2, carrying out HDR fusion reconstruction; s3, Fourier transform is carried out on the arc light periodic noise, and an improved Retinex algorithm is combined to eliminate non-uniform illumination influence; s4, deep learning denoising based on physical simulation; s5, multi-modal data fusion and welding defect decision making are carried out; s6, performing real-time processing and incremental learning; according to the invention, application of artificial intelligence in welding defect identification is realized, the detection accuracy and efficiency are improved, the welding defect cost is reduced, and the intelligent development of welding quality detection is promoted.

The invention relates to the technical field of image processing, and discloses an underwater image enhancement method based on a Retinex algorithm, and the method comprises the steps: firstly obtaining an underwater original image, and extracting multi-dimensional feature information through a preprocessing module (including a histogram analysis unit, a filtering analysis unit and the like); and multi-scale processing is supported, a target processing scale is detected during multi-scale simultaneous processing, and the feature confidence is calculated to screen the optimal feature. The Retinex algorithm processing unit adopts a multi-branch fusion structure (color correction, contrast adjustment and detail recovery branch), generates coefficients and fuses the coefficients to realize enhancement, and can dynamically adjust weights and parameters according to coefficient differences. The method comprehensively extracts features, performs adaptive multi-scale processing, balances the enhancement effect, improves the underwater image quality, and is suitable for underwater image optimization.

The invention discloses an intelligent automobile part quality detection method based on machine vision, and relates to the related field of machine vision and quality detection, and the method comprises the steps: S1, collecting an automobile part surface image through an image collection system; s2, a multi-scale Retinex algorithm based on guided filtering is adopted to pre-process the surface image of the automobile part, and the image quality is improved; s3, using a YOLOv11 instance segmentation model to carry out intelligent detection on the preprocessed automobile part surface image, and identifying and positioning a defect area of the image; and S4, obtaining a defect detection result of the surface of the automobile part, and generating a quality detection report. The problem that in the prior art, the quality detection precision of automobile parts is low is solved. According to the invention, the improved image enhancement algorithm is adopted to improve the image quality, and the learning ability of the neural network is utilized to improve the stability and robustness of quality detection. According to the intelligent quality detection method, the maintenance work of automobile part production is more targeted, and the maintenance and repair cost is effectively reduced.

The invention discloses an electronic detonator bridge wire welding quality on-line detection method and system, and the method comprises the steps: dynamically collecting a visible light and near-infrared dual-channel image, matching a track speed through an encoder, and compensating a pixel; registering, fusing and preprocessing the image, and extracting contour, texture and geometric morphology features; the quality is judged by using an improved ResNet-50 model, and unqualified products are positioned and marked; detection data are classified and stored to realize tracing; the system comprises an image acquisition unit, a preprocessing unit, a feature extraction unit, a quality judgment unit, a positioning marking unit and a classified storage unit, and the image quality is improved through multispectral imaging and dynamic frame rate matching in combination with phase correlation registration and an improved Retinex algorithm; feature extraction is enhanced by means of an attention mechanism, and accurate classification and positioning of defects are realized; a multi-level data mapping structure is established, efficient tracing is supported, the detection efficiency and accuracy are greatly improved, and the high-speed detection requirement of an automatic production line is met.

The invention relates to a citrus surface defect detection method based on a Retinex algorithm and deep learning, and the method comprises the following steps: collecting citrus images of different defect types under different illumination conditions, and marking defect regions and defect types; adjusting the illumination uniformity and detail contrast of the citrus image by using a Retinex algorithm to form a Retinex enhanced image; introducing an attention mechanism of Transform, and performing feature extraction and attention modeling in combination with a convolutional neural network; predicting illumination distribution of the Retinex enhanced image through a deep learning model, and correcting illumination deviation; noise or artifacts in the Retinex enhanced image are removed, and an image after dynamic preprocessing is obtained; and selecting a deep learning model for defect detection, and training a defect detection model by using the dynamically preprocessed image. According to the method, the traditional advantages of image enhancement and the adaptive ability of deep learning are combined, an efficient and accurate solution is provided for citrus surface defect detection, the problems of uneven illumination, shadow interference and the like are effectively solved, and the robustness of defect detection is improved.

The invention relates to the technical field of image processing, in particular to an identity recognition method and system based on image processing. When identity recognition is carried out, image illumination compensation operation is carried out on a collected image through a closed-loop process of semantic perception, type classification and mode-based compensation fusion, the type classification is divided into a low-light scene and an overexposure scene, and for the low-light scene, the overexposure scene is divided into an overexposure scene and an overexposure scene; for an overexposure scene, performing image enhancement processing on the pre-screened face image by adopting a face region distribution guided Retinex algorithm, and for the overexposure scene, pertinently adjusting cliplimit parameters of a CLAHE algorithm by sensing gradient features of a highlight region and a key texture region to realize image enhancement processing on the pre-screened face image; according to the invention, the method achieves the precise adaptation of the face structure, retains the illumination restoration of key details, solves the global defects of an existing method, and lays a high-quality image foundation for the subsequent shielding restoration and posture correction.

The invention provides a diamond wire saw steel ball detection method based on improved yov8. The diamond wire saw steel ball detection method comprises the following steps: S1, acquiring an original image of a diamond wire saw bead; labImage label images are used for marking beads, and data of a training set, a verification set and a test set are formed according to the proportion; s2, carrying out the preprocessing of the image, carrying out the zooming of the image, and carrying out the preprocessing of the image through employing an improved Retinex algorithm; s3, the yov8 model is improved, and an illumination invariant feature extractor (LIF) is added into the backbone; a CA attention mechanism is integrated in a C2f module, and a small target detection head is added and a large-scale detection head is deleted based on small bead size change, so that the small target detection head is used for detecting a bead small target; and S4, optimizing a loss function of the yolov8, improving the small target detection precision by using an improved IoU loss function, and training by using an improved yolov8 model to obtain a training result best.pt file.

The invention relates to the technical field of image processing, in particular to an automobile screw surface corrosion defect detection method, which comprises the following steps of: obtaining a suspected corrosion area according to the local gray scale feature of each pixel point in a gray scale image of a target screw; obtaining a first corrosion confidence coefficient according to the texture characteristics and the gray level distribution complexity of the suspected corrosion area; acquiring a second corrosion confidence coefficient according to the similarity between other suspected corrosion areas in other screws and the suspected corrosion areas and the edge ambiguity of the suspected corrosion areas; according to the first corrosion confidence coefficient and the second corrosion confidence coefficient, obtaining a self-adaptive Gaussian kernel of each pixel point in each suspected corrosion region, and according to the self-adaptive Gaussian kernel of each pixel point in each suspected corrosion region, enhancing the grayscale image by using a Retinex algorithm, and then performing corrosion detection on the target screw. And the detection accuracy of the rusted area on the surface of the automobile screw is improved.

The application provides a kind of underwater structure apparent disease identification method and system based on deep learning, wherein, method includes: according to the combination of fusion model and target recognition model, generate the pre-set underwater structure apparent disease identification model;Underwater apparent disease is identified by the pre-set underwater structure apparent disease identification model;The fusion model is built by improved CycleGAN model and multi-scale Retinex algorithm (MSR) network, for converting underwater image into clear and clear image with obvious features;The target recognition model is obtained by YOLOv5 model, for realizing the positioning and classification of underwater structure apparent disease.The present application solves the problems of inaccurate classification and low recognition accuracy of underwater structure apparent disease by the underwater structure apparent disease identification model, caused by factors such as camera imaging blur, insufficient contrast, dispersion and noise in complex water area.

The present application relates to the technical field of image processing, in particular to an automatic snow removal method and system based on image enhancement processing, comprising the following steps: collecting an original snow image of a road surface, decomposing the image brightness component by using a multi-scale Retinex algorithm, decomposing into a low-frequency component and a high-frequency component, and generating an enhanced image; based on the enhanced image, calculating the joint gradient of saturation and brightness in the pixel point HSV space, and generating a snow region binary segmentation image by using an improved Otsu double threshold method; determining the snow compactness level according to the texture uniformity; driving an execution mechanism to spray snow remover according to path coordinates, and adjusting the snow remover spraying concentration in combination with the compactness level. The present application improves the snow boundary definition and the internal texture recognizability of patches, lays a high-quality image foundation for subsequent segmentation and analysis, and enhances the adaptability and stability of the system under natural light conditions.

The invention relates to an intelligent restoration method for handwriting archive fonts, and the method comprises the steps: decomposing a handwriting archive target image, obtaining channel images corresponding to all color channels, for each channel image, determining a character enhancement coefficient of each pixel point in a character region based on the image feature value of each pixel point in the character region of the channel image, and restoring the character enhancement coefficient of each pixel point in the character region. And then, based on each character enhancement coefficient, correcting a preset Gaussian convolution scale parameter to obtain a target Gaussian convolution scale parameter corresponding to each pixel point in each channel image, and based on a Retinex algorithm and each target Gaussian convolution scale parameter, carrying out character enhancement processing on the corresponding channel image. According to the method, channel images subjected to character enhancement processing are obtained, the channel images subjected to character enhancement processing are combined, the character enhancement image after the target image of the handwritten file is restored is obtained, and restoration of the target image of the handwritten file is achieved.

The invention discloses an image illumination anomaly processing method and device, electronic equipment and a storage medium, which are used for solving the technical problem that the processing effect and the real-time performance cannot be considered at the same time in the prior art. Comprising the following steps: acquiring a brightness component of an RGB image to construct a single-channel brightness matrix; marking an overexposure area and a low-illumination area of the single-channel brightness matrix and calculating a global illumination intensity index; adjusting bilateral filtering parameters according to the global illumination intensity index to adjust a single-channel brightness matrix, and generating a filtered brightness matrix; carrying out enhancement through a Retinex algorithm to obtain an enhanced brightness matrix; determining a key area according to the enhanced brightness matrix and generating a local processing strategy; performing two-layer wavelet transform on the RGB image to obtain a plurality of frequency band components; adjusting each frequency band component according to the global illumination intensity index and the local processing strategy; reconstructing and adjusting the frequency band component through wavelet inverse transformation to obtain a reconstructed RGB image; and converting the reconstructed RGB image into a target image in a preset format.

The invention relates to the technical field of computer vision, in particular to a material defect detection method for keyboard and keycap production, and the method comprises the steps: obtaining an initial image and a grayscale image of a keyboard and a keycap to be detected; obtaining a suspected defect area in the gray level image according to the gray level difference of pixel points in the gray level image; for any suspected defect area, acquiring a defect coefficient of each pixel point in the any suspected defect area according to color features of pixel points of a corresponding area in the initial image and distribution features and texture features of the pixel points of the any suspected defect area; obtaining a self-adaptive small-scale weight and a self-adaptive large-scale weight of each pixel point by using a defect coefficient of each pixel point in the gray level image; according to the self-adaptive small-scale weight and the self-adaptive large-scale weight of each pixel point, an enhanced image is obtained by using a Retinex algorithm, and the accuracy of material defect detection of keyboard and keycap production is improved.

The invention relates to the technical field of image enhancement, and provides a road potential safety hazard monitoring system based on municipal engineering, and the system comprises an image obtaining module which obtains a municipal road image; the image analysis module obtains the area where the municipal road is located according to the gray level distribution; obtaining a sliding detection window of each pixel point in an area where the municipal road is located; acquiring dark particle pixel points in the sliding detection window according to the gray level distribution and the edge features; obtaining the discrete distribution degree of the dark particles according to the position distribution of the dark particle pixel points; obtaining the maximum width moment of the closed edge in the sliding detection window and the defect coefficient of each side length according to the distance between the edge lines; the municipal road surface defect degree is obtained; and the image enhancement module is used for improving a retinex algorithm by combining the municipal road surface defect degree and the discrete distribution degree so as to realize municipal road image enhancement. The invention aims to improve the definition of municipal road image enhancement, so that the details of the municipal road image are more prominent.

The application discloses a camera calibration method and system under uneven illumination conditions, and belongs to the technical field of image processing, and comprises the following steps: S1, processing an input image by using a multi-scale Retinex algorithm; S2, performing Gaussian homomorphic filtering processing on the image obtained in the step S1; S3, processing the image obtained in the step S2 by using an Otsu algorithm; and S4, performing camera calibration on the image processed in the steps S1-S3 by using Zhang Zhengyou plane calibration method. The application firstly enhances image details by using the multi-scale Retinex algorithm and the Gaussian homomorphic filtering, then obtains a final calibration board image by combining Otsu algorithm image segmentation and bilateral filtering processing, and further realizes high-precision calibration of the camera, and is worth popularization and use.

The present invention relates to the field of image data processing technology, and in particular to a tube wall defect detection method for sea wave tube processing. The method comprises: obtaining the grayscale value of a pixel point in the surface image of the sea wave tube; for a target pixel point in the pixel points in the surface image, determining the texture complexity of the reference area of ​​the target pixel point; using edge detection to identify the surface image and obtain all edge pixel points in the surface image; determining the position importance of the reference area of ​​the target pixel point; determining the filtering requirement of the target pixel point; obtaining the corrected Gaussian kernel side length of the target pixel point; using a single-scale Retinex algorithm to enhance the surface image, and then detecting the enhanced surface image through an artificial neural network. The present invention can better process image areas with different features by dynamically adjusting the size of the Gaussian kernel, thereby improving the processing effect of the sea wave tube surface image and thus improving the accuracy of the entire tube wall defect detection.

The invention relates to a metal membrane piece position correction method and system based on improved template matching. The method comprises the steps that an adaptive Retinex algorithm is combined with a DnCNN denoising network to carry out image preprocessing; fusing an ASIFT algorithm and a SuperPoint network to extract feature points and a confidence map; initial matching is realized by using the FREAK descriptor and the weighted Hamming distance, and initial mismatching is filtered based on a bidirectional nearest neighbor criterion; designing a double-layer filtering mechanism in which Guided-PROSAC and local topological structure verification are combined to eliminate mismatching; mapping angular points of the template image to a target image to calculate a pose error vector; the pose error vector is transmitted to a mechanical arm controller for metal diaphragm piece position correction; according to the method, the real-time requirement of a production line is met through reasoning on edge equipment, multiple defects of the traditional technology in the aspects of robustness, precision and automatic integration are overcome, and reliable technical support is provided for a precise assembly task in intelligent manufacturing.

The application discloses a night highway pit and pond recognition method and system based on low-illumination enhancement and parallax calculation, and relates to the technical field of intelligent traffic monitoring and road maintenance. The method comprises the following steps: S1: collecting environmental data, calculating a meteorological index to determine a laser compensation power, and obtaining a stereo image pair to generate environmental metadata; S2: determining an enhancement strategy based on the environmental metadata, and obtaining an enhanced image pair by using an improved Retinex algorithm; S3: performing stereo matching on the enhanced image pair, and generating a scene depth map by fusing pose data; S4: based on a double-branch recognition model, fusing image and depth features, detecting and verifying a pit and pond area, and obtaining a pit and pond area mask; and S5: based on the pit and pond area mask, the depth map and the pose data, calculating three-dimensional parameters and a position of the pit and pond, and generating a detection report. Through adaptive image enhancement and multi-source data fusion technology, the accuracy of pit and pond recognition and the three-dimensional measurement precision in a night low-illumination environment are improved, and automatic and efficient inspection without affecting traffic is realized.

The invention relates to the technical field of image data processing, in particular to a tube wall defect detection method for hypotube processing. Comprising the following steps: acquiring gray values of pixel points in a surface image of a hypotube; for a target pixel point in the pixel points in the surface image, determining texture complexity of a reference area of the target pixel point; identifying the surface image by utilizing edge detection to obtain all edge pixel points in the surface image; determining the position importance degree of the reference area of the target pixel point; determining a filtering demand degree of the target pixel point; obtaining the corrected Gaussian kernel side length of the target pixel point; a surface image is enhanced by using a single-scale Retinex algorithm, and then the enhanced surface image is detected through an artificial neural network. By dynamically adjusting the size of the Gaussian kernel, image areas with different features can be better processed, the processing effect on the hypotube surface image is improved, and therefore the accuracy of defect detection of the whole tube wall is improved.