Image processing method for high-voltage cable defects

Abstract

The invention belongs to the technical field of image processing, and particularly relates to an image processing method for high-voltage cable defects. The method comprises the following steps: firstly, performing multi-scale Gaussian smoothing and other processing on an acquired cable surface real-time image to solve the problem of uneven illumination, and enhancing a brightness dynamic range and local details; afterwards, the image resolution and details are improved through residual learning and frequency domain high-frequency enhancement; performing multi-scale edge detection on the to-be-detected image to obtain edge features; and finally, optimizing edge details through adaptive edge enhancement and noise suppression. The method effectively improves the accuracy and reliability of high-voltage cable defect detection, has remarkable advantages in image processing efficiency, precision and stability, and meets the requirements of high-voltage cable production on surface defect image processing.

Description

Technical Field

[0001] The invention belongs to the technical field of image processing, and in particular relates to an image processing method for high-voltage cable defects. Background Art

[0002] In modern power systems, high-voltage cables play a key role, and their safe and stable operation is crucial to the reliability of power transmission. With the development of technology, image processing technology has received widespread attention and application in the field of high-voltage cable defect detection. In the process of high-voltage cable defect detection, it is a basic link to obtain images of the cable surface through various image acquisition devices. However, the actual acquisition environment is complex and changeable, resulting in many problems with the acquired images. Uneven illumination is one of the common challenges. Uneven illumination can make some areas of the image too bright or too dark, reduce the contrast and clarity of the image, and make it difficult to clearly present the defect features on the cable surface, increasing the difficulty of subsequent defect identification and analysis. In addition, the resolution of the original acquired image may be limited and cannot meet the needs of accurate detection of minor defects. Some minor damage on the cable surface, such as tiny scratches and wear, may be ignored due to insufficient resolution, thus burying safety hazards. At the same time, image edge details are critical for accurately judging the shape, size and position of defects, but in the existing technology, the extraction and enhancement of edge features are often unsatisfactory, which may cause edge blur or loss of important edge information. Summary of the invention

[0003] The present invention aims at the technical problems existing in the image processing of high-voltage cable defects and proposes an image processing method for high-voltage cable defects.

[0004] In order to achieve the above object, the technical solution adopted by the present invention comprises the following steps:

[0005] S1. Collect real-time images of the cable surface to generate images to be tested; process the uneven illumination problem in the images to be tested to enhance the dynamic range of brightness and local details. The specific implementation steps are as follows:

[0006] S11, first input image Use multi-scale Gaussian smoothing to separate the reflection component and the illumination component, and get an image with uniform illumination. The formula is: ,in, To improve the pixel values ​​of images processed by multi-scale Retix, is the contrast enhancement parameter for each size, is a multi-scale weight, satisfying , is a two-dimensional Gaussian kernel function with a scale parameter of ;

[0007] S12, secondly, based on adaptive illumination distribution compensation, the uniformity of light is enhanced to obtain a compensated image ,in For large-scale light distribution, is the global brightness adjustment coefficient;

[0008] S13, finally and Fusion, compressed dynamic range: ,in is the maximum value of the image after illumination homogenization;

[0009] S2. Then, the pixel values ​​of the acquired image need to be processed to improve the image resolution and details through improved residual mapping and frequency domain adaptive high frequency enhancement technology;

[0010] S3, then perform multi-scale edge detection on the image to be detected to obtain edge features at different scales, specifically: ,in represents the multi-scale edge response, is the Laplace edge response at scale k, are adaptive weights of different scales, used to control the contribution of each scale to the total edge response;

[0011] S4. Finally, the multi-scale edges are further optimized using adaptive edge enhancement technology to enhance edge details, and noise suppression processing is applied. The formula is: ,in To control the image gradient The coefficient of strength, is the enhancement factor, Represents the local noise design of the image and completes the processing of the detection image.

[0012] Preferably, the specific implementation of step S2 to improve the image resolution and details is as follows:

[0013] S21, first input low-resolution image Perform residual learning to obtain high-resolution residual : ,in is the target high-resolution image predicted by the neural network. The high resolution is obtained by adding the residual to the low resolution: ,in is the high-resolution image after residual mapping;

[0014] S22, then perform frequency domain high frequency enhancement on the residual result, enhance the high frequency components in the frequency domain through Fourier transform, improve the image edge details, and finally obtain the frequency domain enhanced image ,in represents the inverse Fourier transform, It is the coordinate of the frequency space, which corresponds to the spatial coordinate (x, y) of the image;

[0015] S23, finally, the residual mapping image and the frequency domain enhanced image are fused by weight to obtain a high-resolution output image ,in is the weight coefficient, ranging from 0 to 1.

[0016] Preferably, the specific implementation of step S22 is as follows:

[0017] S221, first perform Fourier transform on the image to obtain frequency domain representation ;

[0018] S222, using an adaptive high-frequency filter to enhance the high-frequency component of the frequency domain image: ,in For high frequency filtering, The range of high-frequency enhancement is dynamically adjusted according to the size and characteristics of the image;

[0019] S223. Finally, the enhanced frequency domain image is converted back to the spatial domain to obtain a frequency domain enhanced image: .

[0020] Preferably, the Laplace operator in step S3 It is used to extract edge features at each scale and adaptively adjust the edge response at each scale through the following formula: ,in is the adjustment coefficient, which controls the sensitivity of weight changes. is the edge enhancement mean, which enables the algorithm to adapt to different defect scales.

[0021] Preferably, the edge enhancement process in step S4 is performed by adding a gradient enhancement term Implemented to enhance edge contrast and suppress local noise To achieve stability, the noise estimation formula is: , where p and q are window sizes, controlling the local extent of the noise estimate.

[0022] Compared with the prior art, the advantages and positive effects of the present invention are that in the treatment of uneven illumination, multi-scale Gaussian smoothing and other technologies are used, fine processing is performed according to different scales, adaptive illumination distribution compensation is performed for reasonable dimming, and fusion compression is performed to ensure suitability, thereby overcoming the defects of traditional means, improving image contrast and clarity, and making defect features clearly presented. For resolution improvement, residual learning and frequency domain high-frequency enhancement technology are used. Residual learning uses low-resolution information, and frequency domain high-frequency enhancement is used for precise processing, avoiding the problem that traditional methods cannot take into account edge details or introduce artifacts, and can accurately detect tiny defects. In terms of edge detection, multi-scale edge detection combined with adaptive edge enhancement and noise suppression technology performs well, can accurately extract edges of defects of different scales, and solves the problems of blurred edges, insensitivity to tiny defects, and susceptibility to noise interference of existing methods. It has obvious advantages in efficiency, accuracy and stability, and meets the requirements of surface defect image processing in high-voltage cable production. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0024] Figure 1 It is a schematic diagram of the method flow of the present invention; DETAILED DESCRIPTION

[0025] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described below in conjunction with the accompanying drawings and embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0026] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways than those described herein. Therefore, the present invention is not limited to the specific embodiments of the following disclosure.

[0027] Embodiment, in the operation and maintenance of high-voltage cables, their safe and stable operation is of great significance to power transmission. Traditional detection methods have gradually exposed many shortcomings when faced with complex actual working conditions. For example, manual inspections are difficult to accurately detect subtle defects and are inefficient. With the development of technology, image processing technology has been applied in cable defect detection, but it still faces challenges. The collected cable images are often affected by uneven lighting. For example, in cable tunnels, the light distribution is complex, resulting in some areas of the image being too bright or too dark, and the defect features are difficult to distinguish. At the same time, the resolution of the original image is limited, and tiny defects are easily overlooked. The edge feature extraction and enhancement effects are not good, which affects the judgment of the defect shape, size and position. To overcome these problems, the present invention aims to provide an effective image processing method to improve the accuracy and reliability of high-voltage cable defect detection and ensure the safe operation of the power system. The overall process is as follows Figure 1 shown.

[0028] First, in order to improve the image contrast and clarity, make the cable surface defect features clearly presented, and solve the problem of uneven illumination causing defects to be difficult to identify, the present invention firstly performs an input image Use multi-scale Gaussian smoothing to separate the reflection component and the illumination component, and get an image with uniform illumination. The formula is: ,in, To improve the pixel values ​​of images processed by multi-scale Retix, is the contrast enhancement parameter for each size, is a multi-scale weight, satisfying , is a two-dimensional Gaussian kernel function with a scale parameter of ; Secondly, based on adaptive illumination distribution compensation, the uniformity of light is enhanced to obtain the compensated image ,in For large-scale light distribution, is the global brightness adjustment coefficient; finally and Fusion, compressed dynamic range: ,in It is the maximum value of the image after the illumination is homogenized; compared with traditional illumination processing methods, it is difficult to adapt to complex illumination, and it is easy to damage local details, making defects difficult to distinguish. The present invention uses multi-scale Gaussian smoothing and other technologies, multi-scale Gaussian smoothing is finely processed according to different scales, adaptive illumination distribution compensation is reasonable dimming, and fusion compression ensures suitability.

[0029] In order to accurately detect tiny defects and meet the needs of detecting minor damage on the cable surface, the present invention then processes the image resolution and details. When traditional methods improve resolution, they often fail to take into account edge details or introduce artifacts. The residual learning and frequency domain high-frequency enhancement technology of the present invention uses low-resolution information for residual learning and accurate processing of frequency domain high-frequency enhancement. The specific processing is to first input a low-resolution image Perform residual learning to obtain high-resolution residual : ,in is the target high-resolution image predicted by the neural network. The high resolution is obtained by adding the residual to the low resolution: ,in is the high-resolution image after residual mapping; then the residual result is enhanced in the frequency domain, the high-frequency component is enhanced in the frequency domain through Fourier transform, and the image is Fourier transformed to obtain the frequency domain representation ; Use an adaptive high-frequency filter to enhance the high-frequency components of the frequency domain image: ,in For high frequency filtering, is the range of high-frequency enhancement, which is dynamically adjusted according to the size and characteristics of the image; finally, the enhanced frequency domain image is converted back to the spatial domain to obtain the frequency domain enhanced image: ,in represents the inverse Fourier transform, It is the coordinate in frequency space, which corresponds to the spatial coordinate (x, y) of the image. Finally, the residual mapping image and the frequency domain enhanced image are fused by weights to obtain a high-resolution output image. ,in is the weight coefficient, ranging from 0 to 1;

[0030] Finally, in order to accurately obtain edge features at different scales and enhance edge details to better judge defect features, multi-scale edge detection and optimization technology is used to perform multi-scale edge detection on the image to be inspected to obtain edge features at different scales. Specifically: ,in represents the multi-scale edge response, is the Laplace edge response at scale k, are adaptive weights of different scales, which are used to control the contribution of each scale to the total edge response. ,in is the adjustment coefficient, which controls the sensitivity of weight changes. is the edge enhancement mean; finally, the adaptive edge enhancement technique is used to further optimize the multi-scale edges to enhance the edge details, and noise suppression is applied. The formula is: ,in To control the image gradient The coefficient of strength, is the enhancement factor, Represents the local noise design of the image and completes the processing of the detection image. , where p and q are window sizes, controlling the local range of noise estimation. Compared with existing edge detection methods, when faced with complex textures and defects of different scales on the cable surface, there may be problems such as blurred edges, insensitivity to minor defects, and susceptibility to noise interference. The multi-scale edge detection of the present invention combines adaptive edge enhancement and noise suppression technology to perform well. Multi-scale edge detection uses the Laplace operator and adaptive weights to accurately extract the edges of defects of different scales. Compared with existing methods, this method has significant advantages in image processing efficiency, accuracy and stability, and meets the requirements of surface defect image processing in high-voltage cable production.

[0031] The above description is only a preferred embodiment of the present invention and does not limit the present invention in other forms. Any technician familiar with the profession may use the technical content disclosed above to change or modify it into an equivalent embodiment with equivalent changes and apply it to other fields. However, any simple modification, equivalent change and modification made to the above embodiment based on the technical essence of the present invention without departing from the content of the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. An image processing method for high voltage cable defects, characterized in that: The following steps are involved: S1. Collect real-time images of the cable surface to generate images to be tested; process the uneven illumination problem in the images to be tested to enhance the brightness dynamic range and local details. The specific implementation steps are as follows: S11, first input image Use multi-scale Gaussian smoothing to separate the reflection component and the illumination component, and get an image with uniform illumination. The formula is: ,in, To improve the pixel values ​​of images processed by multi-scale Retix, is the contrast enhancement parameter for each size, is a multi-scale weight, satisfying , is a two-dimensional Gaussian kernel function with a scale parameter of ; S12, secondly, based on adaptive illumination distribution compensation, the uniformity of light is enhanced to obtain a compensated image ,in For large-scale light distribution, is the global brightness adjustment coefficient; S13, finally and Fusion, compressed dynamic range: ,in is the maximum value of the image after illumination homogenization; S2. Then, the pixel values ​​of the acquired image need to be processed to improve the image resolution and details through improved residual mapping and frequency domain adaptive high frequency enhancement technology; S3, then perform multi-scale edge detection on the image to be detected to obtain edge features at different scales, specifically: ,in represents the multi-scale edge response, is the Laplace edge response at scale k, are adaptive weights of different scales, used to control the contribution of each scale to the total edge response; S4. Finally, the multi-scale edges are further optimized using adaptive edge enhancement technology to enhance edge details, and noise suppression processing is applied. The formula is: ,in To control the image gradient The coefficient of strength, is the enhancement factor, Represents the local noise design of the image and completes the processing of the detection image.

2. The image processing method for high voltage cable defects according to claim 1, characterized in that: The specific implementation of step S2 to improve image resolution and details is as follows: S21, first input low-resolution image Perform residual learning to obtain high-resolution residual : ,in is the target high-resolution image predicted by the neural network. The high resolution is obtained by adding the residual to the low resolution: ,in is the high-resolution image after residual mapping; S22, then perform frequency domain high frequency enhancement on the residual result, enhance the high frequency components in the frequency domain through Fourier transform, improve the image edge details, and finally obtain the frequency domain enhanced image ,in represents the inverse Fourier transform, It is the coordinate of the frequency space, which corresponds to the spatial coordinate (x, y) of the image; S23, finally, the residual mapping image and the frequency domain enhanced image are fused by weight to obtain a high-resolution output image ,in is the weight coefficient, ranging from 0 to 1.

3. The image processing method for high voltage cable defects according to claim 2, characterized in that: The specific implementation of step S22 is: S221, first perform Fourier transform on the image to obtain frequency domain representation ; S222, using an adaptive high-frequency filter to enhance the high-frequency component of the frequency domain image: ,in For high frequency filtering, The range of high-frequency enhancement is dynamically adjusted according to the size and characteristics of the image; S223. Finally, the enhanced frequency domain image is converted back to the spatial domain to obtain a frequency domain enhanced image: .

4. The image processing method for high voltage cable defects according to claim 1, characterized in that: In step S3, the Laplace operator It is used to extract edge features at each scale and adaptively adjust the edge response at each scale through the following formula: ,in is the adjustment coefficient, which controls the sensitivity of weight changes. is the edge enhancement mean, which enables the algorithm to adapt to different defect scales.

5. The image processing method for high voltage cable defects according to claim 1, characterized in that: The edge enhancement process in step S4 is performed by adding a gradient enhancement term Implemented to enhance edge contrast and suppress local noise To achieve stability, the noise estimation formula is: , where p and q are window sizes, controlling the local extent of the noise estimate.

Images