Underwater concrete crack detection method and device based on image processing

Abstract

The invention relates to an underwater concrete crack detection method and device based on image processing, electronic equipment and a medium, and belongs to the technical field of image crack detection.The method comprises the steps that an underwater surface crack image of a target dam is obtained; performing first preprocessing on the surface crack image to obtain a processed to-be-detected surface crack image; inputting the to-be-detected surface crack image into a preset threshold segmentation model to obtain a first image; preprocessing the first image for the second time to obtain a second image; and extracting crack characteristics in the second image, and determining the underwater surface crack grade of the target dam based on the crack characteristics. According to the method, the preprocessed crack image is input into the threshold segmentation model to obtain the first image, and then the first image is preprocessed, so that the crack features are extracted, and the accuracy of crack extraction is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of image crack detection, and in particular to an underwater concrete crack detection method, device, electronic equipment and medium based on image processing. Background Art

[0002] Dams are the most basic facilities in water conservancy and hydropower projects. The part of the dam concrete panel below the water level will inevitably crack due to long-term high water pressure, water scouring, penetration and erosion. The cracks on the dam surface will extend to the inside, causing a surge effect, and then triggering dangerous situations. Cracks in the dam will affect the safe operation of water conservancy projects. Therefore, automatic and accurate detection of underwater cracks in the dam is of great significance for diagnosing dangerous situations, strengthening and repairing, ensuring the safe operation of reservoir dams, and assessing and preventing safety hazards.

[0003] In the prior art, researchers have begun to invest in underwater crack detection of dams and proposed some methods to improve the accuracy of underwater surface crack detection of dams. Padmavathi et al. used homomorphic filtering, anisotropic diffusion filtering, wavelet filtering and other methods, and Celik et al. used Gaussian mixture model to perform nonlinear data mapping on the input image. These methods can increase the contrast of the image and improve the image clarity, but ignore the transmittance, which is easy to cause image oversaturation and distortion, and cannot handle the problem of uneven image illumination. Fan Xinnan et al. proposed a dam crack image enhancement algorithm that imitates underwater biological vision, an artificial bee colony algorithm based on Labor operator, and a rough set dam crack detection method. Shi Dan et al. proposed an underwater image enhancement algorithm based on non-subsampled Cont-ourlet transform and multi-scale Retinex. These methods are mainly suitable for atmospheric lighting environments and can also enhance underwater images, but are easily disturbed by underwater environmental noise, and color cast and other problems will occur, which have great limitations in underwater crack image enhancement processing. Prabhakar et al. integrated four algorithms, namely homomorphic filtering, wavelet denoising, bilateral filtering and contrast equalization, according to the degradation order of underwater images, and proposed an underwater image enhancement method based on image preprocessing, which solved the problems of non-uniform illumination, low contrast and blur of underwater images. The underwater degraded image restoration algorithm based on wavelet transform proposed by Li Qingzhong et al. solved the image blurring and uneven illumination problems. Although the above methods can solve the blurring and uneven illumination problems, they cannot enhance images for specific targets of interest. Fu Jun et al. proposed a dam crack detection algorithm based on BP neural network, which analyzes the brightness value of a two-dimensional image and maps it to a three-dimensional surface image. When the image background changes or the crack features in the image are small and diverse, most detection algorithms cannot effectively and accurately extract the cracks in the image.

[0004] In summary, the prior art lacks a method for accurately extracting cracks from underwater crack images of dams. Summary of the invention

[0005] In view of this, it is necessary to provide an underwater concrete crack detection method, device, electronic device and medium based on image processing to solve the problem of inaccurate crack extraction in underwater crack images of dams in the prior art.

[0006] In order to solve the above problems, the present invention provides an underwater concrete crack detection method based on image processing, comprising: Acquire surface crack images of the target dam underwater; Performing a first preprocessing on the surface crack image to obtain a processed surface crack image to be detected; Inputting the surface crack image to be detected into a preset threshold segmentation model to obtain a first image; Performing a second preprocessing on the first image to obtain a second image; The crack features in the second image are extracted, and the surface crack grade of the target dam underwater is determined based on the crack features.

[0007] In a possible implementation manner, the first preprocessing of the surface crack image includes: The surface crack image is subjected to grayscale conversion, contrast enhancement and filtering denoising processing.

[0008] In a possible implementation, inputting the surface crack image to be detected into a preset threshold segmentation model to obtain a first image includes: The preset threshold segmentation model calculates the local threshold of each area of ​​the surface crack image to be detected according to the brightness distribution of different areas of the surface crack image to be detected; Based on the local threshold of each area of ​​the surface crack image to be detected, a first image is obtained.

[0009] In a possible implementation manner, performing a second preprocessing on the first image to obtain a second image includes: The first image is subjected to morphological processing and connected domain denoising to obtain a second image.

[0010] In a possible implementation manner, performing morphological processing and connected domain denoising on the first image to obtain the second image includes: Performing morphological processing on the first image to obtain a processed first image; The connected domain is denoised according to the roundness of the cracks in the first image to obtain the second image.

[0011] In a possible implementation manner, the crack features in the second image include: one or more of area, length, and width; and extracting the crack features in the second image includes: Acquire a pixel resolution of the second image; Acquire the pixel length, pixel width and pixel area of ​​the crack in the second image; Determining the length of the crack according to the pixel length of the crack and the pixel resolution; Determining a crack width according to the crack pixel width and the pixel resolution; The area of ​​the crack is determined according to the pixel area of ​​the crack and the pixel resolution.

[0012] In a possible implementation, the calculation formula of the pixel resolution is:

[0013] In the formula, J Indicates pixel resolution; L Indicates object distance; f Indicates the focal length of the lens; A Indicates the actual physical size of the crack; Indicates the number of pixels for crack imaging; d Indicates the physical size of the long side of the crack in the sensor; D Indicates the number of pixels on the long side of the crack in the sensor.

[0014] On the other hand, the present invention also provides an underwater concrete crack detection device based on image processing, comprising: A data acquisition module, used to acquire an underwater surface crack image of a target dam; A preprocessing module, used for performing a first preprocessing on the surface crack image to obtain a processed surface crack image to be detected; A first image determination module, used for inputting the surface crack image to be detected into a preset threshold segmentation model to obtain a first image; A second image determination module, used for performing a second preprocessing on the first image to obtain a second image; The crack grade determination module is used to extract crack features in the second image and determine the surface crack grade of the target dam underwater based on the crack features.

[0015] On the other hand, the present invention also provides an electronic device, including a memory and a processor, wherein: The memory is used to store programs; The processor is coupled to the memory and is used to execute the program stored in the memory to implement the steps in the underwater concrete crack detection method based on image processing described in any one of the above implementations.

[0016] On the other hand, the present invention also provides a computer-readable storage medium for storing computer-readable programs or instructions, which, when executed by a processor, can implement the steps of an underwater concrete crack detection method based on image processing described in any of the above-mentioned implementation methods.

[0017] The beneficial effects of the present invention are as follows: the present invention provides an underwater concrete crack detection method based on image processing, firstly obtains the surface crack image of the target dam underwater, and pre-processes the surface crack image to obtain the processed surface crack image to be detected, then inputs the processed surface crack image to be detected into a preset threshold segmentation model to obtain a first image, further pre-processes the first image for a second time to obtain a second image, and finally extracts the crack features in the second image, and determines the surface crack level of the target dam underwater based on the crack features. The present invention inputs the pre-processed surface crack image into the threshold segmentation model to obtain the first image, pre-processes the first image again to obtain the second image, and extracts the crack features from the second image, thereby improving the accuracy of underwater crack extraction of the dam. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 A flowchart of an embodiment of an underwater concrete crack detection method based on image processing provided by the present invention; Figure 2 A schematic diagram of an image grayscale image outputted by a Gamma correction method in an embodiment of an underwater concrete crack detection method based on image processing provided by the present invention; Figure 3 For the present invention Figure 1 A schematic flow chart of an embodiment of S103; Figure 4 A schematic structural diagram of an embodiment of an underwater concrete crack detection device based on image processing provided by the present invention; Figure 5 A schematic structural diagram of an embodiment of an electronic device provided by the present invention. DETAILED DESCRIPTION

[0019] The preferred embodiments of the present invention are described in detail below in conjunction with the accompanying drawings, wherein the accompanying drawings constitute a part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not used to limit the scope of the present invention.

[0020] The present invention provides an underwater concrete crack detection method, device, electronic equipment and medium based on image processing, which are respectively described below.

[0021] Figure 1 A schematic flow chart of an embodiment of the method for detecting underwater cracks in a dam provided by the present invention is shown in FIG. Figure 1 As shown, the underwater crack detection method of the dam includes: S101, acquiring an underwater surface crack image of a target dam; S102, performing a first preprocessing on the surface crack image to obtain a processed surface crack image to be detected; S103, inputting the surface crack image to be detected into a preset threshold segmentation model to obtain a first image; S104, performing a second preprocessing on the first image to obtain a second image; S105: Extract crack features in the second image, and determine the underwater surface crack grade of the target dam based on the crack features.

[0022] Compared with the prior art, the present embodiment provides an underwater concrete crack detection method based on image processing, which first obtains a surface crack image of the target dam underwater, and preprocesses the surface crack image to obtain a processed surface crack image to be detected, and then inputs the processed surface crack image to be detected into a preset threshold segmentation model to obtain a first image, further preprocesses the first image for a second time to obtain a second image, and finally extracts crack features in the second image, and determines the surface crack level of the target dam underwater based on the crack features. The present invention inputs the preprocessed surface crack image into the threshold segmentation model to obtain a first image, preprocesses the first image again to obtain a second image, and extracts crack features from the second image, thereby improving the accuracy of underwater crack extraction of the dam.

[0023] In some embodiments of the present invention, the first preprocessing of the surface crack image includes: The surface crack image is subjected to grayscale conversion, contrast enhancement and filtering denoising processing.

[0024] In a specific embodiment of the present invention, the surface crack image is grayed, specifically: the three-channel RGB image mode of the surface crack image is converted into a single-channel gray image mode, and the image is mainly represented by a brightness index. The drawing paper of the gray image can be quantized into 256 levels, where 0 represents black and 255 represents white. Each gray image can be expressed using a third-order tensor, and the gray value of each pixel can be calculated using the weighted mean of the RGB three-component brightness. The specific calculation formula is as follows, where the gray range of the image gray matrix is ​​0-255.

[0025]

[0026] in, is the grayscale pixel value, , , They are the pixel components in the red, green and blue color channels of the color image respectively. 0.299, 0.587 and 0.114 are the weights. The selection of the weights is based on the human eye's sensitivity to color. Based on these weights, a grayscale image that best matches the human eye's visual perception can be obtained.

[0027] Grayscale the surface crack image. Specifically, after the surface crack image is grayscale transformed, invalid grayscale areas will appear in the crack image, which is not conducive to the subsequent image segmentation processing. Therefore, it is necessary to correct the image to enhance the contrast between the crack and the background and further highlight the crack information. The image Gamma correction method is used to make nonlinear changes to the grayscale value to achieve effective grayscale adjustment. The calculation formula is as follows:

[0028] In the formula, x refers to the input image, γ refers to the power function exponent, F(x) Refers to the output image. γ When it is less than 1, the low gray value area changes more, while the high gray value area changes less, which can enhance the contrast between the two areas. The cracks in the image are low gray value areas, and the background is high gray value areas, which contain noise, such as Figure 2 As shown, take γ The image is corrected when the value is 0.6. The image contrast after correction is high, and the noise target is not significantly enhanced. The correction effect is ideal.

[0029] Filter and denoise the surface crack image. Specifically, the image after contrast enhancement has various noises, including salt and pepper noise and Gaussian noise. Therefore, it is necessary to do a good job of filtering and denoising to eliminate interference information, improve recognition accuracy, and extract crack feature parameters. The median filter algorithm is used to remove the salt and pepper noise on the image to achieve the denoising effect. The formula of the median filter is as follows:

[0030] in, f(x,y) Represents the output pixel, taking the median grayscale value of the pixel in the neighborhood, h(m,n) Represents the grayscale value of the pixel at the center of the original image. m represents the size of the neighborhood window, h(mp,mq) It is the grayscale value of each pixel.

[0031] The bilateral filtering algorithm is used to remove Gaussian noise. The output pixel value of the bilateral filter is the weighted sum of the neighboring pixel values. The weighting coefficient has two parts, one is the spatial proximity factor, and the other is the brightness similarity factor. The formulas are as follows: In the formula, Two pixels and x The distance between phases. In the formula, is the standard deviation between the current pixel and a neighboring pixel. The weighting coefficient of the bilateral filter is a nonlinear combination of these two factors. The spatial proximity factor decreases as the Euclidean distance between the pixel and the center point increases, and the brightness similarity factor decreases as the difference between the brightness values ​​of the two pixels increases. In areas where the image changes slowly, the brightness values ​​of the pixels in the neighborhood are not much different, and the bilateral filter is converted into a Gaussian low-pass filter. In areas where the image changes dramatically, the filter uses the average brightness value of the pixels with similar brightness values ​​near the edge point instead of the original brightness value. Therefore, bilateral filtering not only smoothes the image, but also protects the edge information of the image, showing a certain degree of adaptability.

[0032] In some embodiments of the present invention, Figure 3 As shown, in step S103, the surface crack image to be detected is input into a preset threshold segmentation model to obtain a first image, including: S301, the preset threshold segmentation model calculates the local threshold of each area of ​​the surface crack image to be detected according to the brightness distribution of different areas of the surface crack image to be detected; S302: Obtain a first image based on a local threshold of each region of the surface crack image to be detected.

[0033] In a specific embodiment of the present invention, the preset threshold segmentation model calculates the local threshold of the surface crack image according to the brightness distribution of different regions of the surface crack image. For the image cracks and background, different thresholds can be adaptively calculated to ensure that the threshold of each pixel of the crack image changes with the change of the surrounding neighborhood pixels. The threshold at each pixel position in the surface crack image is not fixed, but is determined by the distribution of the surrounding neighborhood pixels. The binarization threshold of the image area with higher brightness is relatively high, while the binarization threshold of the image area with low brightness will be correspondingly reduced, and the local image areas with different brightness, contrast, and texture will have corresponding local binarization thresholds. This is a good way to overcome the problems of uneven underwater illumination, high noise, contrast, etc., effectively extract the crack area of ​​the underwater image of the dam, and meet the crack detection requirements. For example, the local adaptive threshold uses the Gaussian weighted sum of the local neighborhood block to determine the threshold. If it is larger than this threshold, the pixel at this position is set to 255, and if it is smaller than this threshold, it is set to 0, so as to achieve local threshold segmentation. Since the image is continuous, the closer the points are to the position to be processed, the closer the relationship is, and the farther the points are, the more distant the relationship is. Therefore, the Gaussian weighting method is used to determine the threshold. The closer the point is, the greater the weight is, and the farther the point is, the smaller the weight is. The weight of the pixels around the point to be determined is obtained by the Gaussian equation according to its distance to the center point.

[0034] It should be noted that after the surface crack image to be detected is input into the preset threshold segmentation model for processing, new noise and edge fractures are inevitably generated, so the first image needs to be further preprocessed. In some embodiments of the present invention, the second image is obtained by performing a second preprocessing on the first image, including: The first image is subjected to morphological processing and connected domain denoising to obtain a second image.

[0035] In a specific embodiment of the present invention, the first image is subjected to morphological processing, specifically: using morphological processing to remove small dots and extremely small connected domains that are obviously not cracks, and to connect small broken links in the crack image and smooth the burrs of the crack image.

[0036] After the first image is processed by morphology, there may be some noise. If the complete crack features need to be extracted, the image needs to be filtered again. At this time, the noise can be filtered out by taking advantage of the fact that the crack has a lower roundness and a larger area than the noise. Since the crack area has good connectivity and the connected domain has a larger area, while the noise area is mostly isolated points with a smaller area, a threshold can be set to regard the connected domain below this threshold as a noise area, and then further denoising can be performed using the size of the connected roundness. Roundness is used to describe the degree to which an area is close to a circle. If the area is closer to a circle, its roundness is closer to 1. The definition of roundness in Halcon is used to calculate the roundness. The calculation formula is as follows:

[0037] in C is the roundness, F is the area of ​​the region, max is the maximum value from the center point of the region to the contour point. The roundness of the noise point is generally larger, and the roundness of the crack area is generally smaller. By denoising the connected domain, the crack area can be extracted relatively completely.

[0038] In some embodiments of the present invention, performing morphological processing and connected domain denoising on the first image to obtain the second image includes: Performing morphological processing on the first image to obtain a processed first image; The connected domain is denoised according to the roundness of the cracks in the first image to obtain the second image.

[0039] In some embodiments of the present invention, the crack features in the second image include: one or more of area, length and width; and extracting the crack features in the second image includes: Acquire a pixel resolution of the second image; Acquire the pixel length, pixel width and pixel area of ​​the crack in the second image; Determining the length of the crack according to the pixel length of the crack and the pixel resolution; Determining a crack width according to the crack pixel width and the pixel resolution; The area of ​​the crack is determined according to the pixel area of ​​the crack and the pixel resolution.

[0040] In some embodiments of the present invention, the calculation formula of the pixel resolution is:

[0041] In the formula, J Indicates pixel resolution; L Indicates object distance; f Indicates the focal length of the lens; A Indicates the actual physical size of the crack; Indicates the number of pixels for crack imaging; d Indicates the physical size of the long side of the crack in the sensor; D Indicates the number of pixels on the long side of the crack in the sensor.

[0042] It should be noted that pixel resolution is the proportional relationship between the actual physical size of an object and the corresponding number of pixels in the camera imaging image. Since the crack image is captured using a high-power optical zoom lens, the image field of view is small, and the angle between the optical axis and the normal of the measured plane is small during shooting. Therefore, it can be assumed that the object points corresponding to all pixel points on the image in actual space are at the same distance from the camera projection center, that is, each pixel position in each crack image has the same pixel resolution.

[0043] In a specific embodiment of the present invention, according to the lens imaging principle, there are:

[0044]

[0045] Where L is the object distance and f is the focal length of the lens. Let A be the actual size of the target, i.e. the actual physical size of the crack, and A' be the imaging size, then:

[0046]

[0047] Substituting in:

[0048] Image size for:

[0049] The pixel resolution is:

[0050] In the formula, is the number of pixels of the image, d is the physical size of the long side of the image sensor, D is the number of pixels on the long side of the image sensor.

[0051] Furthermore, based on the obtained pixel resolution, the actual length of the crack, the actual width of the crack and the actual area of ​​the crack are calculated.

[0052] In order to better implement an underwater concrete crack detection method based on image processing in an embodiment of the present invention, based on an underwater concrete crack detection method based on image processing, correspondingly, Figure 4 As shown, an embodiment of the present invention further provides an underwater concrete crack detection device based on image processing. An underwater concrete crack detection device based on image processing 400 includes: The data acquisition module 401 is used to acquire the surface crack image of the target dam underwater; A preprocessing module 402 is used to perform a first preprocessing on the surface crack image to obtain a processed surface crack image to be detected; A first image determination module 403 is used to input the surface crack image to be detected into a preset threshold segmentation model to obtain a first image; A second image determination module 404, configured to perform a second preprocessing on the first image to obtain a second image; The crack grade determination module 405 is used to extract crack features in the second image, and determine the surface crack grade of the target dam underwater based on the crack features.

[0053] The dam underwater crack detection device 400 provided in the above embodiment can implement the technical solution described in the above embodiment of an underwater concrete crack detection method based on image processing. The specific implementation principles of the above modules or units can refer to the corresponding contents in the above embodiment of an underwater concrete crack detection method based on image processing, which will not be repeated here.

[0054] like Figure 5 As shown, the present invention also provides an electronic device 500. The electronic device 500 includes a processor 501, a memory 502 and a display 503. Figure 5 Only some components of the electronic device 500 are shown, but it should be understood that it is not required to implement all of the components shown, and more or fewer components may be implemented instead.

[0055] In some embodiments, the processor 501 may be a central processing unit (CPU), a microprocessor or other data processing chip, used to run program codes or process data stored in the memory 502, such as an underwater concrete crack detection method based on image processing in the present invention.

[0056] In some embodiments, the processor 501 may be a single server or a server group. The server group may be centralized or distributed. In some embodiments, the processor 501 may be local or remote. In some embodiments, the processor 501 may be implemented in a cloud platform. In some embodiments, the cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an internal cloud, a multi-cloud, etc., or any combination thereof.

[0057] In some embodiments, the memory 502 may be an internal storage unit of the electronic device 500, such as a hard disk or memory of the electronic device 500. In other embodiments, the memory 502 may also be an external storage device of the electronic device 500, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the electronic device 500.

[0058] Furthermore, the memory 502 may include both an internal storage unit of the electronic device 500 and an external storage device. The memory 502 is used to store application software installed in the electronic device 500 and various data.

[0059] In some embodiments, the display 503 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch device, etc. The display 503 is used to display information on the electronic device 500 and to display a visual user interface. The components 501-503 of the electronic device 500 communicate with each other via a system bus.

[0060] In one embodiment, when the processor 501 executes an underwater concrete crack detection program based on image processing in the memory 502, the following steps may be implemented: Acquire surface crack images of the target dam underwater; Performing a first preprocessing on the surface crack image to obtain a processed surface crack image to be detected; Inputting the surface crack image to be detected into a preset threshold segmentation model to obtain a first image; Performing a second preprocessing on the first image to obtain a second image; The crack features in the second image are extracted, and the surface crack grade of the target dam underwater is determined based on the crack features.

[0061] It should be understood that: when the processor 501 executes an underwater concrete crack detection program based on image processing in the memory 502, in addition to the above functions, other functions can also be realized. For details, please refer to the description of the corresponding method embodiment above.

[0062] Furthermore, the embodiment of the present invention does not specifically limit the type of the electronic device 500 mentioned, and the electronic device 500 may be a portable electronic device such as a mobile phone, a tablet computer, a personal digital assistant (PDA), a wearable device, a laptop computer, etc. Exemplary embodiments of portable electronic devices include but are not limited to portable electronic devices equipped with IOS, Android, Microsoft or other operating systems. The above-mentioned portable electronic devices may also be other portable electronic devices, such as a laptop computer with a touch-sensitive surface (e.g., a touch panel). It should also be understood that in some other embodiments of the present invention, the electronic device 500 may not be a portable electronic device, but a desktop computer with a touch-sensitive surface (e.g., a touch panel).

[0063] Those skilled in the art will appreciate that all or part of the processes of the above-mentioned embodiments can be implemented by instructing related hardware through a computer program, and the program can be stored in a computer-readable storage medium, wherein the computer-readable storage medium is a disk, an optical disk, a read-only storage memory, or a random access memory, etc.

[0064] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by any technician familiar with the technical field within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention.

Claims

1. An underwater concrete crack detection method based on image processing, characterized in that: include: Acquire surface crack images of the target dam underwater; Performing a first preprocessing on the surface crack image to obtain a processed surface crack image to be detected; Inputting the surface crack image to be detected into a preset threshold segmentation model to obtain a first image; Performing a second preprocessing on the first image to obtain a second image; The crack features in the second image are extracted, and the surface crack grade of the target dam underwater is determined based on the crack features.

2. The underwater concrete crack detection method based on image processing according to claim 1 is characterized in that: The first preprocessing of the surface crack image comprises: The surface crack image is subjected to grayscale conversion, contrast enhancement and filtering denoising processing.

3. The underwater concrete crack detection method based on image processing according to claim 1 is characterized in that: The step of inputting the surface crack image to be detected into a preset threshold segmentation model to obtain a first image includes: The preset threshold segmentation model calculates the local threshold of each area of ​​the surface crack image to be detected according to the brightness distribution of different areas of the surface crack image to be detected; A first image is obtained based on a local threshold of each region of the surface crack image to be detected.

4. The underwater concrete crack detection method based on image processing according to claim 1 is characterized in that: The performing a second preprocessing on the first image to obtain a second image includes: The first image is subjected to morphological processing and connected domain denoising to obtain a second image.

5. The underwater concrete crack detection method based on image processing according to claim 1 is characterized in that: The step of performing morphological processing and connected domain denoising on the first image to obtain the second image includes: Performing morphological processing on the first image to obtain a processed first image; The connected domain is denoised according to the roundness of the cracks in the first image to obtain the second image.

6. The underwater concrete crack detection method based on image processing according to claim 1 is characterized in that: The crack features in the second image include: one or more of area, length and width; and extracting the crack features in the second image includes: Acquire a pixel resolution of the second image; Acquire the pixel length, pixel width and pixel area of ​​the crack in the second image; Determining the length of the crack according to the pixel length of the crack and the pixel resolution; Determining a crack width according to the crack pixel width and the pixel resolution; The area of ​​the crack is determined according to the pixel area of ​​the crack and the pixel resolution.

7. The underwater concrete crack detection method based on image processing according to claim 1 is characterized in that: The calculation formula of the pixel resolution is: In the formula, J Indicates pixel resolution; L Indicates object distance; f Indicates the focal length of the lens; A Indicates the actual physical size of the crack; Indicates the number of pixels for crack imaging; d Indicates the physical size of the long side of the crack in the sensor; D Indicates the number of pixels on the long side of the crack in the sensor.

8. An underwater concrete crack detection device based on image processing, characterized in that: include: A data acquisition module, used to acquire an underwater surface crack image of a target dam; A preprocessing module, used for performing a first preprocessing on the surface crack image to obtain a processed surface crack image to be detected; A first image determination module, used for inputting the surface crack image to be detected into a preset threshold segmentation model to obtain a first image; A second image determination module, used for performing a second preprocessing on the first image to obtain a second image; The crack grade determination module is used to extract crack features in the second image and determine the surface crack grade of the target dam underwater based on the crack features.

9. An electronic device, characterized in that: comprising a memory and a processor, wherein: The memory is used to store programs; The processor is coupled to the memory and is used to execute the program stored in the memory to implement the steps of the underwater concrete crack detection method based on image processing as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: Used to store computer-readable programs or instructions, which, when executed by a processor, can implement the steps of the underwater concrete crack detection method based on image processing as described in any one of claims 1 to 7.

Images