Road anticorrosion effect detection and evaluation method based on image processing technology

Abstract

The invention relates to the technical field of image processing, in particular to a road anti-corrosion effect detection and evaluation method based on an image processing technology, and the method comprises the steps: obtaining to-be-detected grayscale images corresponding to two different types of images on the surface of a road anti-corrosion layer, and carrying out the segmentation to obtain to-be-detected sub-images; respectively obtaining two difference images corresponding to the to-be-detected sub-image according to the row gray difference and the column gray difference of the gray difference image corresponding to the to-be-detected sub-image; obtaining the roughness degree of the to-be-detected sub-images, obtaining an optimal scale based on the roughness degree, decomposing the two to-be-detected grayscale images by using wavelet transform according to the optimal scale, and fusing according to decomposition results to obtain a fused feature image; and obtaining a defect region of the fused feature image, and obtaining a road quality detection result according to area features of the defect region. According to the invention, a more accurate road quality detection result can be obtained.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of image processing, in particular to a highway anticorrosion effect detection and evaluation method based on image processing technology. BACKGROUND

[0002] With the deepening of the development and utilization of marine resources, underwater highways as a key underwater facility sealing device, its role is increasingly prominent. Underwater highways are mainly used for sealing underwater pipelines or wellheads, which can effectively prevent pipeline leaks and avoid environmental pollution, and protect the safe and stable operation of underwater facilities. Given the important role of underwater highways in marine engineering, quality detection has become a key link to ensure facility safety.

[0003] Under the current technical conditions, the quality detection of underwater highways mainly focuses on the corrosion of the anticorrosion layer, because the integrity of the anticorrosion layer directly affects the anticorrosion performance and service life of the highway. Therefore, detecting the changes in the anticorrosion layer has become one of the routine quality detection items. The existing detection method usually relies on the surface images of the highway anticorrosion layer collected by the camera, and evaluates the quality of the highway by analyzing the corrosion area in the image.

[0004] However, this method has certain limitations in practical application. Due to the complexity of the underwater environment and the limitations of camera imaging, the collected surface images often contain a lot of noise, which will interfere with the accurate extraction of the corrosion area. The presence of noise may lead to misjudgment or omission of the corrosion area, thereby affecting the accuracy of the quality detection result.

[0005] In order to improve the accuracy of the corrosion detection of the anticorrosion layer of the underwater highway, a new detection and evaluation method based on image processing technology needs to be developed. This method should be able to effectively remove image noise and accurately extract the corrosion area, so as to realize accurate evaluation of the anticorrosion effect of the underwater highway. This will help to timely discover and handle the problems of the anticorrosion layer, ensure the safe and stable operation of the underwater highway, and prolong its service life. SUMMARY

[0006] In order to solve the technical problem that the quality detection result of the highway is not accurate, the purpose of the present application is to provide a highway anticorrosion effect detection and evaluation method based on image processing technology, and the technical scheme adopted is as follows: Obtain two different types of images corresponding to the detected gray scale image of the surface of the highway anticorrosion layer, and segment the detected gray scale image to obtain the detected sub-image; obtain two difference images corresponding to the detected sub-image according to the row gray scale difference and column gray scale difference of the gray scale difference image corresponding to the detected sub-image; Obtain the roughness of the detected sub-image according to the difference of the pixel value and the texture information in the two difference images corresponding to the detected sub-image; The preset scale of wavelet transform is adjusted by using the roughness to obtain the optimal scale. The two grayscale images to be detected are decomposed by wavelet transform according to the optimal scale, and then the decomposition results are fused to obtain the fused feature image. The defect area of ​​the fused feature image is obtained, and the quality inspection result of the highway is obtained according to the area characteristics of the defect area.

[0007] Preferably, the two difference images corresponding to the sub-image to be detected are obtained according to the row grayscale difference and column grayscale difference of the grayscale difference image corresponding to the sub-image to be detected, specifically: For any sub-image to be detected, the absolute value of the difference between the grayscale values ​​of the pixels at corresponding positions in adjacent rows in the corresponding grayscale difference image is calculated to obtain a first difference image of the sub-image to be detected; the absolute value of the difference between the grayscale values ​​of the pixels at corresponding positions in adjacent columns in the corresponding grayscale difference image is calculated to obtain a second difference image of the sub-image to be detected; The difference image includes a first difference image and a second difference image.

[0008] Preferably, the roughness of the sub-image to be detected is obtained according to the difference in pixel values ​​and texture information in the two difference images corresponding to the sub-image to be detected, specifically: For any difference image corresponding to the sub-image to be detected, the variance of the pixel values ​​of all pixels is recorded as a roughness coefficient; threshold segmentation is performed on the difference image to obtain a first feature region, where the pixel values ​​of the pixels in the first feature region are greater than a preset threshold; and the overlapping area of ​​the first feature regions of the two difference images is obtained to obtain a second feature region; A first roughness weight and a second roughness weight are obtained according to texture information of the second feature area, and the roughness coefficients of the first difference image and the second difference image are weightedly summed using the first roughness weight and the second roughness weight to obtain the roughness corresponding to the sub-image to be detected.

[0009] Preferably, obtaining the first roughness weight and the second roughness weight according to the texture information of the second feature area is specifically: The second feature area is processed using a vertical filter and a horizontal filter respectively to obtain a vertical response result and a horizontal response result; For the response results in any direction, the mean and standard deviation of the phase response values ​​are obtained, and the rough weight is obtained according to the mean and standard deviation of the phase response values. The mean is positively correlated with the rough weight, and the standard deviation is negatively correlated with the rough weight. The rough weight corresponding to the response result in the horizontal direction is used as the first rough weight, and the rough weight corresponding to the response result in the vertical direction is used as the second rough weight.

[0010] Preferably, the method for obtaining the gray difference image corresponding to the to-be-detected sub-image is specifically as follows: For any to-be-detected sub-image, a center pixel point of the to-be-detected sub-image is obtained, if the number of the center pixel points is even, the mean value of the pixel values of all the center pixel points is taken as the feature pixel value, if the number of the center pixel points is one, the pixel value of the center pixel point is taken as the feature pixel value; the difference between the pixel value of each pixel point in the to-be-detected sub-image and the feature pixel value is calculated respectively to obtain a gray difference image.

[0011] Preferably, the adjusting the preset scale of the wavelet transform by using the roughness to obtain an optimal scale is specifically as follows: For any to-be-detected sub-image, the sum value between the preset value and the roughness of the to-be-detected sub-image is obtained, the ratio between the preset scale of the wavelet transform and the sum value is calculated, and the optimal scale corresponding to the to-be-detected sub-image is obtained by rounding the ratio.

[0012] Preferably, the to-be-detected gray image corresponding to the two different types of images of the surface of the highway anticorrosive layer is specifically as follows: The RGB image and the infrared image of the surface of the highway anticorrosive layer are obtained respectively, the RGB image is processed to obtain a first gray image, and the infrared image is processed to obtain a second gray image; the first gray image and the second gray image are to-be-detected gray images.

[0013] Preferably, the fusion feature image is obtained according to the decomposition result, and the method is specifically as follows: The decomposition result includes the high-frequency coefficient and the low-frequency coefficient of the to-be-detected gray image. The maximum value of the high-frequency coefficients of the first gray image and the second gray image is taken as a reconstructed high-frequency coefficient, and the low-frequency coefficient information of the first gray image and the second gray image is weighted and summed by using a preset weight to obtain a reconstructed low-frequency coefficient. The fusion feature image is reconstructed by using the reconstructed high-frequency coefficient and the reconstructed low-frequency coefficient for inverse wavelet decomposition.

[0014] Preferably, the quality detection result of the highway is obtained according to the area characteristics of the defect area, and the method is specifically as follows: The area of each defect area is obtained, and the ratio between the sum value of the areas of all the defect areas and the total area of the fusion feature image is taken as a quality coefficient; when the quality coefficient is less than a preset first threshold value, the quality detection result of the corresponding highway anticorrosive layer is qualified; when the quality coefficient is greater than or equal to the first threshold value and less than a preset second threshold value, the quality detection result of the corresponding highway anticorrosive layer is a first-level damage; when the quality coefficient is greater than or equal to the second threshold value, the quality detection result of the corresponding highway anticorrosive layer is a second-level damage; the first threshold value is less than the second threshold value.

[0015] The embodiments of the present application have at least the following beneficial effects: The present application firstly acquires the to-be-detected gray images corresponding to two different types of images, so that subsequent feature fusion can be performed based on the decomposition results of the two different types of images, a more accurate feature image can be obtained, and then the to-be-detected gray image is uniformly segmented to obtain a to-be-detected sub-image, so that subsequent analysis of local pixel changes of the to-be-detected gray image can be performed, and two difference images of each to-be-detected sub-image are obtained, that is, the gray differences in two directions are analyzed respectively, and the roughness of the to-be-detected sub-image is obtained in combination with the texture information, so that the image feature extraction result is more accurate. Then, the preset scale of wavelet transform is adjusted by using the roughness to obtain an optimal scale, so that the scale parameter in the wavelet transform can be adaptively obtained when the to-be-detected gray image is decomposed, so that the decomposition result is more optimal, and then a fusion feature image with more optimal edges and detail textures is obtained. Based on the fusion feature image, the defect area is extracted, and then the quality detection result of the underwater highway anticorrosive coating is more accurate. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, and the advantages thereof, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained without creative labor based on these drawings.

[0017] Figure 1 is a method flowchart of the highway anticorrosion effect detection and evaluation method based on image processing technology of the present application. DETAILED DESCRIPTION

[0018] In order to further illustrate the technical means and effects adopted by the present application to achieve the predetermined invention purpose, the following will combine the drawings and the preferred embodiments to specifically describe the specific implementation, structure, features and effects of the highway anticorrosion effect detection and evaluation method based on image processing technology according to the present application. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs.

[0020] The specific scheme of the highway anticorrosion effect detection and evaluation method based on image processing technology provided by the present application will be specifically described below in combination with the drawings.

[0021] The main purpose of the present application is: by shooting the RGB image and the infrared image of the underwater highway anticorrosive layer surface, optimizing the preset scale in wavelet change based on the local roughness in the image, so that different scales can be adaptively obtained, the corresponding gray images of the two different types of images are respectively decomposed and fused, and finally the quality detection result of the highway anticorrosive layer is obtained based on the image obtained after fusion.

[0022] Embodiment: Please refer to Figure 1 , which shows the method flow chart of the highway anticorrosive effect detection and evaluation method based on image processing technology provided by an embodiment of the present application, and the method comprises the following steps: Step one, obtaining two gray images of the highway anticorrosive layer surface to be detected, segmenting the gray images to be detected to obtain sub-images to be detected; obtaining two difference images corresponding to the sub-images to be detected according to the row gray difference and column gray difference of the gray difference images corresponding to the sub-images to be detected.

[0023] Firstly, the professional underwater camera equipment is used to shoot the RGB image and the infrared image of the highway anticorrosive layer. When shooting the image underwater, the collected image may be affected by noise, so it is necessary to carry out denoising processing on the collected image. In this embodiment, the median filter is used to carry out denoising processing on the collected image, and the process is a known technology, which will not be introduced in detail here. The RGB image is subjected to gray processing to obtain a first gray image, and the infrared image is subjected to gray processing to obtain a second gray image; the first gray image and the second gray image are gray images to be detected. There are many methods for gray processing of the image, and in this embodiment, the maximum value method is used for gray processing of the image. The implementer can select according to the specific implementation scene.

[0024] The first gray image corresponding to the RGB image of the highway anticorrosive layer surface has a good presentation effect on the texture information and detail edges of each part of the object in the image, but there is environmental noise such as bubbles, suspended matter, etc. when collecting the image, which will have a certain influence on the imaging effect of the RGB image. The infrared image of the highway anticorrosive layer surface has a large tolerance to environmental noise and can better present the characteristics of the observed object.

[0025] The infrared radiation results of the corroded part and the non-corroded part of the highway anticorrosive coating exist certain differences, and then if the highway anticorrosive coating exists corrosion, the corrosion characteristics of the highway anticorrosive coating can be observed in the second gray-scale image corresponding to the infrared image, and the influence of environmental factors is smaller. However, the second gray-scale image corresponding to the infrared image has poor presentation effect on the texture information and detailed edges of each object in the image, and therefore, in the embodiment, the feature information in the first gray-scale image and the second gray-scale image is analyzed respectively, the first gray-scale image and the second gray-scale image are fused, that is, the parts with better presentation effect are fused, so as to avoid the influence of underwater environmental noise, and at the same time, more accurate and clear detailed information can be obtained, and then the more accurate corrosion area in the fused image can be extracted.

[0026] In the embodiment, the wavelet transform algorithm is used to perform multi-scale decomposition on two different types of gray-scale images to be detected on the surface of the highway anticorrosive coating, and low-frequency coefficients and high-frequency coefficients are obtained respectively, different fusion methods are used for different coefficients, and then the fused image is obtained through wavelet decomposition inverse transformation. When the traditional wavelet transform is used to decompose the image, the scale is a fixed value, and the decomposition result is poor, and therefore, in the embodiment, the preset scale in the wavelet decomposition transformation algorithm is optimized by analyzing the texture information and other features in the gray-scale image to be detected, so that the image can be decomposed based on the adaptive scale when the image is decomposed, and the decomposition result is better.

[0027] Based on this, in order to better analyze the detailed information and texture features in the gray-scale image to be detected, the gray-scale image to be detected needs to be segmented, that is, the local features in the gray-scale image to be detected are analyzed, and the optimal scale corresponding to the local area in the gray-scale image to be detected is obtained when the wavelet decomposition is performed.

[0028] In the embodiment, the gray-scale image to be detected is uniformly segmented to obtain a detected sub-image, that is, each detected sub-image is uniformly segmented into a preset number of regions with the same size, and the image of each region is used as a detected sub-image, wherein the value of the preset number is determined by the size of the gray-scale image to be detected by the implementer, for example, if the size of the gray-scale image to be detected is 16x16, the gray-scale image to be detected is uniformly segmented to obtain four regions with the size of 4x4.

[0029] Further, the pixel value change of the pixel point in the to-be-detected sub-image is analyzed, that is, two difference images corresponding to the to-be-detected sub-image are obtained according to the row gray difference and the column gray difference of the gray difference image corresponding to the to-be-detected sub-image. In this embodiment, the first gray image corresponding to the RGB image and the second gray image corresponding to the infrared image are both subjected to uniform segmentation processing, and then the gray difference image of each to-be-detected sub-image in the first gray image and the second gray image needs to be obtained respectively.

[0030] The method for obtaining the gray difference image is specifically as follows: for any to-be-detected sub-image, the center pixel points of the to-be-detected sub-image are obtained, if the number of the center pixel points is even, the average of the pixel values of all the center pixel points is taken as the feature pixel value, if the number of the center pixel points is one, the pixel value of the center pixel point is taken as the feature pixel value; the difference between the pixel value of each pixel point in the to-be-detected sub-image and the feature pixel value is calculated to obtain the gray difference image.

[0031] For example, if the size of the to-be-detected sub-image is 5*5, the pixel point located at the center of the to-be-detected sub-image is the center pixel point, at this time, there is only one center pixel point, so the pixel value of the center pixel point can be taken as the feature pixel value. If the size of the to-be-detected sub-image is 6*6, the pixel point located at the center of the to-be-detected sub-image is the center pixel point, at this time, there are four center pixel points, so the average of the pixel values of the four center pixel points can be taken as the feature pixel value.

[0032] Further, the difference between the pixel value of each pixel point in the to-be-detected sub-image and the center pixel point is analyzed, for example, for the first pixel point in the to-be-detected sub-image, the absolute value of the difference between the pixel value of the first pixel point and the feature pixel value is calculated, and then the absolute value of the difference between each pixel point in the to-be-detected sub-image and the feature pixel point is calculated as the pixel value at the corresponding position in the gray difference image.

[0033] The gray difference image can represent the change of the pixel point in the corresponding to-be-detected sub-image relative to the center of the region, if there is no defect part in the to-be-detected sub-image, the gray difference between each pixel point in the to-be-detected sub-image and the center pixel point of the region is small, if there is a defect part in the to-be-detected sub-image, the relative difference between each pixel point in the to-be-detected sub-image and the center pixel point of the region can be obtained, and then the gray difference change in the image is analyzed.

[0034] For any one of the to-be-detected sub-image, the absolute value of the difference between the gray values of the corresponding position pixel points in the adjacent rows of the corresponding gray difference image is calculated to obtain the first difference image of the to-be-detected sub-image; the absolute value of the difference between the gray values of the corresponding position pixel points in the adjacent columns of the corresponding gray difference image is calculated to obtain the second difference image of the to-be-detected sub-image; the difference image includes the first difference image and the second difference image.

[0035] For example, the absolute value of the difference between the gray values of the corresponding position pixel points in the first row and the second row of the gray difference image is calculated, and then the absolute value of the difference between the gray values of the corresponding position pixel points in the second row and the third row is calculated, and so on, until the difference between the gray values of the corresponding position pixel points in every two adjacent rows is calculated, to obtain the first difference image of the to-be-detected sub-image. The absolute value of the difference between the gray values of the corresponding position pixel points in the first column and the second column of the gray difference image is calculated, and then the absolute value of the difference between the gray values of the corresponding position pixel points in the second column and the third column is calculated, and so on, until the difference between the gray values of the corresponding position pixel points in every two adjacent columns is calculated, to obtain the second difference image of the to-be-detected sub-image.

[0036] The first difference image can reflect the change characteristics of the gray difference between each pixel point in the to-be-detected sub-image and the center pixel point in the horizontal direction, and the second difference image can reflect the change characteristics of the gray difference between each pixel point in the to-be-detected sub-image and the center pixel point in the vertical direction.

[0037] Step two, obtaining the roughness of the to-be-detected sub-image according to the difference between the pixel values in the two difference images corresponding to the to-be-detected sub-image and the texture information.

[0038] Each to-be-detected gray image corresponds to a plurality of to-be-detected sub-images, and each to-be-detected sub-image corresponds to two difference images, including the first difference image and the second difference image. When there is a defect part in the to-be-detected sub-image, the gray difference between each pixel point in the to-be-detected sub-image and the center pixel point may change significantly, so the feature area with large gray difference change can be obtained according to the change of the pixel value in the difference image.

[0039] Based on this, for any one of the difference images corresponding to the to-be-detected sub-image, the first feature area is obtained by threshold segmentation of the difference image, and the pixel value of the pixel point in the first feature area is greater than the preset threshold; the second feature area is obtained by obtaining the overlapping area of the first feature areas of the two difference images.

[0040] In this embodiment, the first difference image is taken as an example for explanation. The maximum inter-class variance method is used to obtain the segmentation threshold in the first difference image as the preset threshold. The area composed of pixels in the first difference image whose pixel values ​​are greater than the preset threshold is used as the first feature area in the first difference image. The first feature area corresponding to the second difference image can be obtained according to the same method. The first feature area in the first difference image represents the part where the grayscale difference of each pixel point relative to the center pixel point in the sub-image to be detected changes greatly in the horizontal direction. The first feature area corresponding to the second difference image represents the part where the grayscale difference between each pixel point relative to the center pixel point in the sub-image to be detected changes greatly in the vertical direction. The overlapping part between the parts with large changes in both directions is recorded as the second feature area. The second feature area can represent the part where the overall grayscale difference of each pixel point relative to the center pixel point in the sub-image to be detected changes greatly.

[0041] For any difference image corresponding to the sub-image to be inspected, the variance of the pixel values ​​across all pixels is recorded as the roughness coefficient. The roughness coefficient corresponding to the first difference image represents the degree of fluctuation in the horizontal grayscale variation of the sub-image to be inspected, while the roughness coefficient corresponding to the second difference image represents the degree of fluctuation in the vertical grayscale variation of the sub-image to be inspected. When a defect is present in the sub-image to be inspected, a larger value of the roughness coefficient in the corresponding difference image indicates a greater degree of fluctuation in the grayscale variation.

[0042] When analyzing the roughness of the corresponding sub-image to be detected based on the roughness coefficients of the first difference image and the second difference image, if the roughness distribution in the sub-image to be detected is more biased towards the vertical direction, the roughness coefficient corresponding to the second difference image needs to be given a greater weight; if the roughness distribution in the sub-image to be detected is more biased towards the horizontal direction, the roughness coefficient corresponding to the first difference image needs to be given a greater weight.

[0043] In this embodiment, the roughness distribution in the sub-image to be detected is characterized by analyzing the texture features of the portion with the largest overall grayscale difference in the sub-image to be detected. That is, the first roughness weight and the second roughness weight are obtained based on the texture information of the second feature area.

[0044] Specifically, taking the second feature region corresponding to any one to-be-detected sub-image as an example, the second feature region is processed by using the filter in the vertical direction and the filter in the horizontal direction respectively to obtain the response result in the vertical direction and the response result in the horizontal direction; for the response result in any one direction, the mean value and the standard deviation of the phase response value are obtained, and the coarse weight is obtained according to the mean value and the standard deviation of the phase response value, the mean value and the coarse weight are in a positive correlation relationship, and the standard deviation and the coarse weight are in a negative correlation relationship; the coarse weight corresponding to the response result in the horizontal direction is taken as the first coarse weight, and the coarse weight corresponding to the response result in the vertical direction is taken as the second coarse weight.

[0045] In the embodiment, the Gabor filter is used to process the second feature region in the to-be-detected sub-image, and the Gabor filter is used to extract the image feature response of the second feature region in different directions and different scales. The Gabor filter is a filter based on the Gabor function, and the Gabor function is a sinusoidal wave with a Gaussian function modulation, which can be used to describe the texture feature of an image. The response result of the Gabor filter is a two-dimensional complex matrix, which represents the response of the texture feature of the input image in different directions and different scales. Specifically, the amplitude of the response result represents the texture feature intensity of the input image in different directions and different scales, and the phase of the response result represents the texture feature distribution direction of the image.

[0046] In order to obtain the texture feature roughness of the second feature region in the to-be-detected sub-image in the horizontal direction and the texture feature roughness in the vertical direction, the second feature region is processed by using the Gabor filter in the vertical direction and the Gabor filter in the horizontal direction respectively to obtain the response result in the vertical direction and the response result in the horizontal direction. That is, the Gabor filter in two directions is used to convolve the second feature region in the to-be-detected sub-image to obtain the phase response map in the vertical direction and the phase response map in the horizontal direction, and then the mean value and the standard deviation of all phase response values in each direction of the phase response map are calculated.

[0047] In the embodiment, taking the second feature region in any one to-be-detected sub-image as an example, the calculation formula of the first coarse weight can be represented as: wherein, α represents the first coarse weight corresponding to the second feature region in the to-be-detected sub-image, represents the mean value of all phase response values in the horizontal direction response result, represents the standard deviation of all phase response values in the horizontal direction response result, and Norm() represents a normalization function, represents a first preset value, and in this embodiment, the value is 0.01. In order to prevent the denominator from being 0, the implementer can set it according to the specific implementation scene.

[0048] the mean value of all phase response values in the horizontal direction response result The greater the value of the first roughness weight, the greater the phase response value in the horizontal direction, and the more concentrated the texture feature distribution in the second feature region in the to-be-detected sub-image is in the horizontal direction, thereby reflecting that the direction of the rough distribution in the second feature region is closer to the horizontal direction, and the greater the value of the first roughness weight.

[0049] the standard deviation of all phase response values in the horizontal direction response result The greater the value of the first roughness weight, the greater the phase response value in the horizontal direction, and the more concentrated the texture feature distribution in the second feature region in the to-be-detected sub-image is in the horizontal direction, thereby reflecting that the direction of the rough distribution in the second feature region is closer to the horizontal direction, and the greater the value of the first roughness weight.

[0050] The second roughness weight can be obtained in the same way as the calculation formula of the first roughness weight. The first roughness weight represents the degree to which the rough texture distribution of the second feature region in the to-be-detected sub-image is concentrated in the horizontal direction, and the second roughness weight represents the degree to which the rough texture distribution of the second feature region in the to-be-detected sub-image is concentrated in the vertical direction. At the same time, the second feature region represents a region with a large change in gray difference in the to-be-detected sub-image, which may be a region corresponding to a defect part, so the texture information corresponding to the second feature region is used to weight the overall rough distribution of the to-be-detected sub-image.

[0051] Specifically, the first roughness weight and the second roughness weight are used to weight and sum the roughness coefficients of the first difference image and the second difference image to obtain the roughness degree corresponding to the to-be-detected sub-image. The calculation formula of the roughness degree corresponding to the to-be-detected sub-image can be represented as: wherein Q represents the roughness degree corresponding to the to-be-detected sub-image, a represents the first roughness weight, represents the roughness coefficient of the first difference image, β represents the second roughness weight, represents the roughness coefficient of the second difference image.

[0052] The greater the value of the first roughness weight, the more concentrated the rough texture distribution in the to-be-detected sub-image is in the horizontal direction, the greater the value of the roughness coefficient of the first difference image, the greater the fluctuation degree of the gray difference change in the to-be-detected sub-image in the horizontal direction, and the greater the value of the corresponding roughness degree, indicating that the to-be-detected sub-image is more likely to have a defect part.

[0053] The greater the value of the second roughness weight, the more concentrated the rough texture distribution in the to-be-detected sub-image is in the vertical direction, the greater the value of the rough coefficient of the second difference image, the greater the fluctuation degree of the gray difference change in the vertical direction of the to-be-detected sub-image, and the greater the value of the roughness, indicating that the to-be-detected sub-image is more likely to have a defect part.

[0054] In step three, the preset scale of the wavelet transform is adjusted by using the roughness to obtain an optimal scale, and the two to-be-detected gray images are decomposed by using the wavelet transform according to the optimal scale, and a fused feature image is obtained according to the decomposition result.

[0055] When the two to-be-detected gray images, i.e., the first gray image and the second gray image, are fused, the texture information and the detail edge effect of the defect part in the image are better in the first gray image, and the main feature effect of the defect part in the image is better in the second gray image. Therefore, the wavelet decomposition transform algorithm is used to analyze the to-be-detected gray images respectively to obtain low-frequency and high-frequency coefficients, different fusion methods are used for different coefficients corresponding to the two to-be-detected gray images, and then the wavelet decomposition inverse transform is used to obtain the fused images of the two to-be-detected gray images, which can better fuse the parts with better representation effects of the two to-be-detected gray images, so that the subsequent defect analysis based on the fused images has a more accurate effect.

[0056] In this embodiment, the discrete wavelet transform and the inverse transform are used to process the first gray image and the second gray image, at the same time, the multi-Besse wavelet basis function is used as the wavelet basis function in the discrete wavelet transform algorithm, and the scale is used as one of the important parameters in the wavelet basis function. The fixed numerical scale makes the decomposition result less effective.

[0057] Based on this, after the surface of the underwater highway anticorrosive layer is corroded, the corresponding to-be-detected gray image presents a dark gray color, and the corroded part may have a concave condition with a high roughness. The normal anticorrosive layer surface without corrosion has a lighter color in the corresponding to-be-detected gray image, and does not have a concave condition with a low roughness. When the local area in the to-be-detected gray image is processed by the discrete wavelet transform, if the local area contains a corrosion defect part, the corresponding roughness is high, a smaller scale is used to make the corresponding wavelet basis function have a larger frequency, and more accurate detail information, i.e., high-frequency information, can be obtained. If the local area does not contain a corrosion defect part, the corresponding roughness is low, a larger scale is used to make the corresponding wavelet basis function have a smaller frequency, and the approximate information, i.e., low-frequency information, in the image can be better represented.

[0058] Based on this, the roughness degree corresponding to the to-be-detected sub-image in the to-be-detected gray image represents the roughness of the texture in the to-be-detected sub-image, and the preset scale of the wavelet transform is adjusted by using the roughness degree to obtain an optimal scale. Specifically, for any to-be-detected sub-image, the sum value between the preset value and the roughness degree of the to-be-detected sub-image is obtained, the ratio between the preset scale of the wavelet transform and the sum value is calculated, and the optimal scale corresponding to the to-be-detected sub-image is obtained by rounding the ratio. In this embodiment, the calculation formula of the optimal scale can be represented as: wherein, represents the optimal scale corresponding to the to-be-detected sub-image, j represents the preset scale, and Q represents the roughness degree of the to-be-detected sub-image. is a down-rounding function, is a preset value, which is 1 in this embodiment. In order to prevent the denominator from being 0, the implementer can set it according to the specific implementation scenario.

[0059] It should be noted that the preset scale is a scale parameter in the wavelet basis function and the scale function in the wavelet decomposition transform, and different scale parameters correspond to different decomposition results. In this embodiment, the value of the preset scale is 5, and the preset scale is adjusted by using the roughness degree, and the adjusted value is processed by using the down-rounding function, so that the final obtained scale value is an integer. The implementer can set it according to the specific implementation scenario.

[0060] The greater the roughness degree of the to-be-detected sub-image, the more rough the surface texture distribution in the to-be-detected sub-image, and thus it is indicated that there may be a defect part in the to-be-detected sub-image, and the corresponding function frequency is higher. The result of decomposition by using the optimal scale is better in representing the detail information in the to-be-detected sub-image. The smaller the roughness degree of the to-be-detected sub-image, the smoother the surface texture in the to-be-detected sub-image, and thus it is indicated that there may be no defect part in the to-be-detected sub-image, and the corresponding function frequency is lower. The result of decomposition by using the optimal scale is better in representing the main part in the to-be-detected sub-image.

[0061] At this point, the adaptive optimal scale corresponding to each to-be-detected sub-image in each to-be-detected gray image can be obtained. Based on the optimal scale and by using the Daubechies wavelet basis function in the discrete wavelet transform algorithm, each to-be-detected sub-image in each to-be-detected gray image is decomposed, and the decomposition result of each to-be-detected gray image is obtained.

[0062] It should be noted that for any one to be detected gray scale image, each to be detected sub-image corresponding to the image exists corresponding area in the to be detected gray scale image, that is, when the to be detected gray scale image is processed by discrete wavelet transform, the corresponding area corresponding to each to be detected sub-image in the to be detected gray scale image is respectively decomposed to obtain the decomposition result of the to be detected gray scale image. The decomposition result of the to be detected gray scale image includes high frequency coefficients and low frequency coefficients of the to be detected gray scale image.

[0063] In the traditional discrete wavelet transform algorithm, two groups of coefficients, namely approximation coefficients and detail coefficients, are output. The approximation coefficients represent the output of the low-pass filter of the wavelet transform, and the detail coefficients represent the output of the high-pass filter of the wavelet transform. In this embodiment, the output approximation coefficients are denoted as low frequency coefficients, and the output detail coefficients are denoted as high frequency coefficients.

[0064] The maximum value of the high frequency coefficients of the first gray scale image and the second gray scale image is obtained as the reconstructed high frequency coefficient, which can maximize the retention of edge and texture information in the image. The low frequency coefficient information of the first gray scale image and the second gray scale image is weighted and summed using a predetermined weight to obtain a reconstructed low frequency coefficient, so that the low frequency optical characteristics affect the fusion result to a greater extent and distortion is avoided. The reconstructed high frequency coefficient and the reconstructed low frequency coefficient are used for wavelet decomposition inverse transform to reconstruct a fusion feature image.

[0065] Specifically, in this embodiment, the first weight corresponding to the low frequency coefficient of the first gray scale image is set to 0.3, and the second weight corresponding to the low frequency coefficient of the second gray scale image is set to 0.7. The implementer can set according to the specific implementation scene, and then the low frequency coefficient information of the first gray scale image and the second gray scale image is weighted and summed using the first weight and the second weight to obtain the reconstructed low frequency coefficient.

[0066] After obtaining the reconstructed high frequency coefficient and the reconstructed low frequency coefficient, the reconstructed high frequency coefficient and the reconstructed low frequency coefficient are used as the decomposition result of the discrete wavelet transform, and then the decomposition result is subjected to discrete wavelet inverse transform to obtain a reconstructed image, which is denoted as a fusion feature image. Discrete wavelet transform and discrete wavelet inverse transform are well-known techniques, and their transformation processes are not described in detail here.

[0067] Step four, obtaining a defect area of the fusion feature image, and obtaining a quality detection result of the highway according to the area characteristics of the defect area.

[0068] The obtained fusion feature image avoids underwater environmental noise as much as possible, and can better highlight the detailed features of the defect part in the image, and thus the defect area in the image can be accurately extracted. Specifically, the implementer can extract the defect area in the fusion feature image through threshold segmentation, connected domain extraction, region growing, and the like. In this embodiment, the region growing algorithm is used to process the fusion feature image.

[0069] Specifically, the pixel point corresponding to the minimum value of the pixel value in the fusion feature image is taken as the initial seed point, and the condition for region growing is that the absolute value of the difference between the pixel values of the pixel points is less than or equal to the pixel threshold value. In this embodiment, the value of the pixel threshold value is 20, and the implementer can set it according to the specific implementation scenario. Further, region growing is performed according to the initial seed point and the region growing condition, and the obtained region is recorded as a defect area.

[0070] Since the corrosion defect of the underwater highway anticorrosive coating is darker in color than the normal part of the anticorrosive coating surface, the defect area obtained through region growing is the corrosion defect area part of the underwater highway anticorrosive coating, and thus the quality detection result of the highway is obtained according to the area feature of the defect area.

[0071] Specifically, the area of each defect area is obtained, and the ratio between the sum of the areas of all defect areas and the total area of the fusion feature image is taken as a quality coefficient. When the quality coefficient is less than a preset first threshold value, the quality detection result of the corresponding highway anticorrosive coating is qualified; when the quality coefficient is greater than or equal to the first threshold value and less than a preset second threshold value, the quality detection result of the corresponding highway anticorrosive coating is a first-level damage; when the quality coefficient is greater than or equal to the second threshold value, the quality detection result of the corresponding highway anticorrosive coating is a second-level damage; and the first threshold value is less than the second threshold value. Among them, the degree of second-level damage is more serious than that of first-level damage.

[0072] In the embodiment, the first threshold value is 0.15, and the second threshold value is 0.3. The implementer can set the values according to the specific implementation scene. When the quality coefficient is less than 0.15, it indicates that there are fewer defect parts on the surface of the underwater highway anticorrosive coating, that is, there is only slight damage, which does not affect the continued use of the anticorrosive coating of the highway, so the quality detection result of the highway anticorrosive coating at this time is qualified. When the quality coefficient is greater than or equal to 0.15 and less than 0.3, it indicates that there are more defect parts on the surface of the underwater highway anticorrosive coating, that is, there is moderate damage, and there is a certain quality problem. The relevant staff needs to take timely measures to avoid safety hazards, so the quality detection result of the highway anticorrosive coating at this time is first-level damage. When the quality coefficient is greater than 0.3, it indicates that there are the most defect parts on the surface of the underwater highway anticorrosive coating, that is, there is serious damage, which indicates that there is a serious quality problem on the surface of the underwater highway anticorrosive coating. The relevant staff needs to take timely measures to avoid safety hazards, so the quality detection result of the highway anticorrosive coating at this time is second-level damage.

[0073] In summary, the quality detection method of the highway anticorrosive coating is improved to avoid the defects of the conventional discrete wavelet transform method. The image processing method is used to first collect the RGB image and the infrared image of the surface of the underwater highway anticorrosive coating, use the discrete wavelet transform to decompose the corresponding gray images of the two images, and use different fusion methods for different frequency coefficients to avoid the influence of the underwater environmental noise and achieve the effect of highlighting the details of the corrosion defect parts. At the same time, the roughness of the to-be-detected sub-image is obtained according to the pixel feature distribution of the to-be-detected sub-image corresponding to the to-be-detected gray image, and the scale parameter of the Daubechies wavelet basis function in the discrete wavelet variation is adaptively adjusted according to the roughness, so that the effect of using the scale parameter for the discrete wavelet transform and the inverse discrete wavelet transform of the image is better, and then a fusion feature image with better edges and texture details is obtained. The defect area is extracted based on the fusion feature image, and then the quality detection result of the underwater highway anticorrosive coating obtained is more accurate.

[0074] The above-described embodiments are only used to illustrate the technical solutions of the present application, but not limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements for some technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. A method for detecting and evaluating highway anti-corrosion effects based on image processing technology, characterized in that: The method comprises the following steps: Obtain grayscale images to be detected corresponding to two different types of images of the highway anti-corrosion layer surface, segment the grayscale images to be detected to obtain sub-images to be detected; and obtain two difference images corresponding to the sub-images to be detected based on the row grayscale differences and column grayscale differences of the grayscale difference images corresponding to the sub-images to be detected; The roughness of the sub-image to be detected is obtained according to the difference in pixel values ​​and texture information in the two difference images corresponding to the sub-image to be detected; The preset scale of wavelet transform is adjusted by using the roughness to obtain the optimal scale. The two grayscale images to be detected are decomposed by wavelet transform according to the optimal scale, and then the decomposition results are fused to obtain the fused feature image. The defect area of ​​the fused feature image is obtained, and the quality inspection result of the highway is obtained according to the area characteristics of the defect area.

2. The highway anti-corrosion effect detection and evaluation method based on image processing technology according to claim 1 is characterized in that: The two difference images corresponding to the sub-image to be detected are obtained according to the row grayscale difference and column grayscale difference of the grayscale difference image corresponding to the sub-image to be detected: For any sub-image to be detected, the absolute value of the difference between the grayscale values ​​of the pixels at corresponding positions in adjacent rows in the corresponding grayscale difference image is calculated to obtain a first difference image of the sub-image to be detected; the absolute value of the difference between the grayscale values ​​of the pixels at corresponding positions in adjacent columns in the corresponding grayscale difference image is calculated to obtain a second difference image of the sub-image to be detected; The difference image includes a first difference image and a second difference image.

3. The highway anti-corrosion effect detection and evaluation method based on image processing technology according to claim 2 is characterized in that: The roughness of the sub-image to be detected is obtained according to the difference in pixel values ​​and texture information in the two difference images corresponding to the sub-image to be detected: For any difference image corresponding to the sub-image to be detected, the variance of the pixel values ​​of all pixels is recorded as a roughness coefficient; threshold segmentation is performed on the difference image to obtain a first feature region, where the pixel values ​​of the pixels in the first feature region are greater than a preset threshold; and the overlapping area of ​​the first feature regions of the two difference images is obtained to obtain a second feature region; A first roughness weight and a second roughness weight are obtained according to texture information of the second feature area, and the roughness coefficients of the first difference image and the second difference image are weightedly summed using the first roughness weight and the second roughness weight to obtain the roughness corresponding to the sub-image to be detected.

4. The method for detecting and evaluating highway anti-corrosion effects based on image processing technology according to claim 3 is characterized in that: The first roughness weight and the second roughness weight are obtained according to the texture information of the second feature area as follows: The second feature area is processed using a vertical filter and a horizontal filter respectively to obtain a vertical response result and a horizontal response result; For the response results in any direction, the mean and standard deviation of the phase response values ​​are obtained, and the rough weight is obtained according to the mean and standard deviation of the phase response values. The mean is positively correlated with the rough weight, and the standard deviation is negatively correlated with the rough weight. The rough weight corresponding to the response result in the horizontal direction is used as the first rough weight, and the rough weight corresponding to the response result in the vertical direction is used as the second rough weight.

5. The highway anti-corrosion effect detection and evaluation method based on image processing technology according to claim 1 is characterized in that: The method for obtaining the grayscale difference image corresponding to the sub-image to be detected is specifically as follows: For any sub-image to be detected, obtain the central pixel point of the sub-image to be detected. If the number of central pixels is even, use the average of the pixel values ​​of all central pixels as the feature pixel value. If the number of central pixels is one, use the pixel value of the central pixel as the feature pixel value. The difference between the pixel value of each pixel in the sub-image to be detected and the feature pixel value is calculated respectively to obtain a grayscale difference image.

6. The highway anti-corrosion effect detection and evaluation method based on image processing technology according to claim 1 is characterized in that: The preferred scale obtained by adjusting the preset scale of the wavelet transform using the roughness is specifically: For any sub-image to be detected, the sum of the preset value and the roughness of the sub-image to be detected is obtained, the ratio between the preset scale of the wavelet transform and the sum is calculated, and the ratio is rounded to obtain the preferred scale corresponding to the sub-image to be detected.

7. The highway anti-corrosion effect detection and evaluation method based on image processing technology according to claim 1 is characterized in that: The grayscale images to be detected corresponding to the two different types of images of the highway anti-corrosion layer surface are specifically: An RGB image and an infrared image of the surface of the highway anti-corrosion layer are obtained respectively, and the RGB image is grayscaled to obtain a first grayscale image, and the infrared image is grayscaled to obtain a second grayscale image; the first grayscale image and the second grayscale image are grayscale images to be detected.

8. The method for detecting and evaluating highway anti-corrosion effects based on image processing technology according to claim 7 is characterized in that: The fusion feature image obtained by fusion according to the decomposition results is specifically: The decomposition result includes high-frequency coefficients and low-frequency coefficients of the grayscale image to be detected; Obtaining the maximum value of the high-frequency coefficients of the first grayscale image and the second grayscale image as the reconstructed high-frequency coefficient, and weighted summing the low-frequency coefficient information of the first grayscale image and the second grayscale image using a preset weight to obtain the reconstructed low-frequency coefficient; The reconstructed high-frequency coefficients and the reconstructed low-frequency coefficients are used to perform inverse wavelet decomposition and transform to obtain the fused feature image.

9. The highway anti-corrosion effect detection and evaluation method based on image processing technology according to claim 1 is characterized in that: The quality inspection results of the highway obtained according to the area characteristics of the defective area are specifically: The area of ​​each defective region is obtained, and the ratio between the sum of the areas of all defective regions and the total area of ​​the fused feature image is used as the quality coefficient. When the quality coefficient is less than a preset first threshold, the quality inspection result of the corresponding highway anti-corrosion layer is qualified; when the quality coefficient is greater than or equal to the first threshold and less than a preset second threshold, the quality inspection result of the corresponding highway anti-corrosion layer is first-level damage; when the quality coefficient is greater than or equal to the second threshold, the quality inspection result of the corresponding highway anti-corrosion layer is second-level damage; The first threshold is smaller than the second threshold.

Images