Coating structure corrosion non-destructive testing method and repair evaluation method

Abstract

The present application relates to a method for non-destructive testing of painted structure corrosion and a method for maintenance evaluation, comprising: presetting a crawling track of a crawler, after each step of the crawler, an ultrasonic probe scans along a direction perpendicular to the step direction, after a full scan, the crawler moves one step further; after each step of the crawler, a local visual detection system on the crawler takes a picture of the area to be scanned by the ultrasonic probe, obtaining a local painted structure surface image; during the movement of the ultrasonic probe, ultrasonic reflection signals are received; rust identification is performed on the local painted structure surface image to obtain a surface rust image recognition result; at the same time, hidden rust identification is performed on the ultrasonic reflection signals to obtain a hidden rust identification result; and the surface rust image recognition result and the hidden rust identification result are fused to obtain a painted structure corrosion detection result. The method realizes accurate evaluation of painted structure corrosion, has high accuracy and is easy to implement.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of nondestructive testing, in particular to a method for nondestructive testing of corrosion of a structure with paint and a method for maintenance evaluation. BACKGROUND

[0002] There are many large structures with paint in the prior art, such as storage tanks and cruise ships. However, these large structures with paint are easily corroded on the surface of the structure when exposed to the natural environment for a long time, which greatly reduces the service life. Therefore, regular maintenance is needed for these large structures with paint, which includes regular inspection, cleaning, and rust removal and corrosion prevention treatment.

[0003] For example, large storage tanks generally adopt the method of periodic tank cleaning and erection of peripheral scaffolding to achieve overall tank rust removal and brushing of anticorrosive paint. This method can achieve good maintenance effect, but the maintenance interval is long, the construction period is long, the cost is high, and regular maintenance cannot be targeted according to the surface rust condition of each tank, which will cause a certain waste of resources. In addition, sometimes the surface anticorrosive paint is removed and the rust area is maintained according to the results of manual visual inspection, but it cannot find the situation that the paint layer is not damaged and rust has occurred under the coating, causing the rust area not to be effectively treated in time. Therefore, how to realize nondestructive testing of the rust condition of the outer wall of the tank body has important practical significance for the safety of the production area.

[0004] Currently, there are various diagnostic techniques for storage tank wall quality, such as infrared thermal imaging detection, eddy current detection, X-ray detection, ultrasonic detection, and visual detection. Each method is based on different detection principles and has its own advantages and disadvantages in actual detection. Among them:

[0005] (1) Infrared thermal imaging detection technology

[0006] Infrared thermal imaging detection technology is based on infrared physics as the theoretical basis, and determines the damage condition of the tank wall according to the influence of the temperature field distribution caused by the different thicknesses of the side wall. The advantages are that it can realize large-area rapid detection of various damage types, has intuitive image display, does not need to contact the workpiece, etc. The disadvantages are that the detection is greatly affected by the temperature factor in the environment, the qualitative determination of defects is relatively difficult, and the quantitative precision of defects is low.

[0007] (2) Eddy current detection technology

[0008] Eddy current testing is based on electromagnetic induction principle, applied to the surface and near-surface detection method of conductive medium, using eddy current probe close to the side wall of the tank, in the wall surface generated induced eddy current, when the tank wall has rust defects, resulting in the phase and amplitude of eddy current change, realize the detection of rust defects. The advantage is that it can directly contact and couple without cleaning the tank, and the sensitivity to surface and near-surface defects is high, but it is difficult to make quantitative judgment of defects due to the interference of surface topography;

[0009] (3) X-ray detection technology

[0010] X-ray detection technology is to use the different absorption rates of tank wall intact area and corrosion area to X-ray, and then show different degrees of radiation attenuation at the receiving end. The advantage is that the corrosion area is intuitive and has high sensitivity, and is not affected by the structure, but the tank needs to be cleaned before detection, and attention should be paid to the X-ray radiation range during detection;

[0011] (4) Ultrasonic testing

[0012] The principle of ultrasonic testing is that ultrasonic waves will produce reflection and scattering phenomena when acting on different media according to the difference in acoustic impedance between different materials or media, and when defects occur in the material, the amplitude and phase of the transmitted ultrasonic waves change, which is used to evaluate the defects. The traditional ultrasonic testing method is mainly used for macroscopic and large-size defects, and the detection cost is moderate. In addition, combined with the positioning information of mechanical devices, C-scan imaging of internal defects can be realized, making the detection results more intuitive.

[0013] (5) Visual inspection

[0014] Visual inspection is one of the simplest and most widely used methods. It usually requires professional inspectors and is greatly affected by subjective factors. It can detect surface corrosion that has already appeared, but it is powerless for corrosion hidden under the paint layer.

[0015] Currently, the detection of tank wall corrosion still relies on ultrasonic testing, which uses ultrasonic velocity and propagation time to calculate the tank wall thickness, and can detect corrosion defects on the tank wall in a timely manner. It has high detection accuracy for large-scale macro defects on the tank wall. For tank wall rust detection, it relies on manual visual inspection, and it is powerless for rust hidden under the paint layer. There is no suitable non-destructive testing method in the industry, and only through periodic maintenance, the surface paint layer can be removed and identified manually. However, due to the differences in rust removal technology, personnel and equipment during routine maintenance and cleaning of the outer wall, it is difficult to ensure the consistency of tank rust removal quality, and it is difficult to predict the internal rust generation and expansion after painting anti-rust paint. Therefore, a non-destructive testing method is needed to non-destructively and quantitatively evaluate the rust condition of the outer wall of the painted structure (tank), especially the rust condition under the paint layer. SUMMARY

[0016] The technical problem solved by the present application is to provide a paint structure corrosion nondestructive detection method capable of automatically realizing rust detection and having high accuracy.

[0017] The technical problem solved by the present application is to provide a paint structure maintenance evaluation method applying the paint structure corrosion nondestructive detection method.

[0018] The first technical solution adopted by the present application to solve the above technical problem is a paint structure corrosion nondestructive detection method, characterized by comprising a global visual detection system and an ultrasonic detection system, the ultrasonic detection system comprising a crawler capable of crawling on the paint structure and an ultrasonic probe and a local visual detection system carried on the crawler.

[0019] The paint structure corrosion nondestructive detection method specifically comprises the following steps:

[0020] Step 1: placing the crawler on the surface of the paint structure to be detected to determine the placement position of the global visual detection system so that the shooting field of view of the global visual detection system can cover the crawling area of the crawler.

[0021] Step 2: starting the global visual detection system, at which time the global visual detection system starts to collect images in the field of view and identifies and locates the crawler in the images to identify the center position of the crawler and the scanning area of the ultrasonic probe.

[0022] Step 3: presetting the crawling track of the crawler, after the crawler crawls one step, the ultrasonic probe scans along the direction perpendicular to the stepping direction, after scanning the whole scanning area, the crawler crawls one step further; after the crawler crawls one step, the local visual detection system on the crawler takes a photo of the scanning area of the ultrasonic probe to obtain a local paint structure surface layer image, and receives the ultrasonic reflection signal during the movement of the ultrasonic probe.

[0023] Step 4: rust identification is performed on the local paint structure surface layer image to obtain a surface rust image identification result, at the same time, concealed rust identification is performed on the ultrasonic reflection signal to obtain a concealed rust identification result, and the surface rust image identification result and the concealed rust identification result are fused to obtain a paint structure corrosion detection result.

[0024] Preferably, a scanning support is further arranged on the crawler, the scanning support is arranged along the direction perpendicular to the advancing direction of the crawler, and the ultrasonic probe is constrained on the scanning support in a manner capable of moving along the extension direction of the scanning support.

[0025] Preferably, the specific steps of identifying the scanning area of the ultrasonic probe in Step 2 are as follows:

[0026] Fix the crawler to the detection starting position of the structure to be detected with paint, and the ultrasonic probe is located at the starting point of the scanning support, start the global visual detection system, and move the ultrasonic probe from the starting point of the scanning support to the end point of the scanning support, acquire the first image when the ultrasonic probe is at the starting point of the scanning support through the global visual detection system, and acquire the second image when the ultrasonic probe is at the end point of the scanning support through the global visual detection system;

[0027] The first image and the second image are respectively subjected to binaryzation processing, and then the binaryzation processed first image and the binaryzation processed second image are subjected to XOR operation, the region with pixel value 1 in the image after XOR operation is the region where the ultrasonic probe is located, the position with pixel value 1 in the binaryzation processed first image corresponds to the starting point position of the ultrasonic probe, and the position with pixel value 1 in the binaryzation processed second image corresponds to the end point position of the ultrasonic probe, so the connecting region between the starting point position of the ultrasonic probe and the end point position of the ultrasonic probe in the region is the scanning region of the ultrasonic probe.

[0028] Preferably, the specific steps of identifying the center position of the crawler are as follows:

[0029] The binaryzation processed first image and the binaryzation processed second image are subjected to AND operation, and the mean values of the horizontal coordinates and the vertical coordinates of all pixel points in the image after AND operation are calculated, that is, the horizontal coordinate and the vertical coordinate of the center position of the crawler are obtained.

[0030] Preferably, the specific process of rust identification on the local paint structure surface image in step 4 is as follows:

[0031] First, the global visual detection system is used to acquire the overall appearance image of the paint structure to be detected, the initial paint color of the paint structure to be detected is determined by color statistics, and the brightness difference of the paint color at each position under the current illumination is realized;

[0032] In the local paint structure surface image acquired by the local visual detection system, the current surface paint color of the paint structure to be detected is RGB(r, g, b), the RGB image of the entire paint structure to be detected is acquired, and the mean value of the paint color RGB and the standard deviation δ are calculated by respectively performing histogram statistics on RGB. Whether it is within ±k*δ, k is a constant greater than 0, if yes, it is determined that the current paint color is a normal color, if no, it is determined that the current paint color is an abnormal color, and the rust area in the paint is segmented according to the normal color and the abnormal color.

[0033] Preferably, the specific process of rust identification on the local paint structure surface image in step 4 is as follows:

[0034] The ultrasonic emission signal includes an interface wave, a bottom wave and noise, whether the maximum amplitude of the interface wave is greater than a times the average amplitude of the noise and the maximum amplitude of the bottom wave is lower than b times the average amplitude of the noise is judged, a and b are both constants, a>b, if yes, it is determined that the area detected by the current ultrasonic probe is rusted under the paint layer; if no, it is determined that the area detected by the current ultrasonic probe is not rusted under the paint layer.

[0035] Preferably, the specific process of fusing the surface rust image recognition result and the hidden rust recognition result in step 4 is:

[0036] The rust area in the hidden rust recognition result is written into the rust area in the surface rust image recognition result, so as to realize the fusion of the surface rust image recognition result and the hidden rust recognition result.

[0037] The second technical solution adopted by the present application to solve the above technical problems is: a paint structure maintenance evaluation method applied to the paint structure corrosion nondestructive detection method, characterized by comprising: calculating the rust proportion of the paint structure, judging whether the rust proportion of the paint structure exceeds a preset maintenance setting value, if yes, the paint structure needs to be maintained; if no, the paint structure does not need to be maintained.

[0038] Compared with the prior art, the present application has the advantages that: the local visual detection system on the crawler takes a picture of the area to be scanned by the ultrasonic probe, and obtains a local paint structure surface image; and the ultrasonic probe receives the ultrasonic reflection signal, so that the visual and ultrasonic detection are fused to realize accurate evaluation of the corrosion condition of the paint structure, which has the advantages of high accuracy and easy implementation. BRIEF DESCRIPTION OF DRAWINGS

[0039] Figure 1 It is a schematic diagram of the ultrasonic detection system in the embodiment of the present application;

[0040] Figure 2 It is another perspective view of the schematic diagram of the ultrasonic detection system in the embodiment of the present application; Figure 1

[0041] Figure 3 It is a schematic diagram of the ultrasonic detection system in the embodiment of the present application for detecting a vertical steel storage tank;

[0042] Figure 4 It is a schematic diagram of the ultrasonic detection system in the embodiment of the present application for detecting a horizontal steel storage tank;

[0043] Figure 5 It is a relationship diagram of the crawler movement track and the ultrasonic probe movement track in the embodiment of the present application;

[0044] Figure 6 ​A schematic diagram of the surface rust image recognition result and the hidden rust recognition result in the embodiment of the present application. DETAILED DESCRIPTION

[0045] The present application will be further described in detail below with reference to the embodiments of the drawings.

[0046] The non-destructive testing method for the painted structure corrosion in the embodiment includes a global visual detection system and an ultrasonic detection system, as shown in Figure 1 and Figure 2 The ultrasonic detection system includes a crawler 3 capable of crawling on the painted structure, and an ultrasonic probe 1 and a local visual detection system 2 carried on the crawler 3; the crawler 3 is further provided with a scanning bracket 4 arranged perpendicular to the advancing direction of the crawler 3, and the ultrasonic probe 1 is constrained on the scanning bracket 4 in a manner capable of moving along the extension direction of the scanning bracket 4. The crawler 3 in the embodiment is a magnetic adsorption crawler, which includes magnetic adsorption rollers 31; the magnetic adsorption crawler is a prior art, and its structure is a prior art, which will not be described here. The global visual detection system and the local visual detection system 2 in the embodiment are both CCD cameras.

[0047] The non-destructive testing method for the painted structure corrosion specifically includes the following steps:

[0048] Step 1, placing the crawler on the surface of the painted structure to be detected, and determining the placement position of the global visual detection system so that the shooting field of view of the global visual detection system can cover the crawling area of the crawler;

[0049] Step 2, starting the global visual detection system, at this time the global visual detection system starts to collect images in the field of view, and identifies and locates the crawler in the images, identifies the center position of the crawler and the ultrasonic probe scanning area;

[0050] The specific steps for identifying the ultrasonic probe scanning area are as follows:

[0051] Fixing the crawler at the detection starting position of the painted structure to be detected, while the ultrasonic probe is at the starting point of the scanning bracket, starting the global visual detection system, and moving the ultrasonic probe from the starting point of the scanning bracket to the end point of the scanning bracket, collecting a first image when the ultrasonic probe is at the starting point of the scanning bracket by the global visual detection system, and collecting a second image when the ultrasonic probe is at the end point of the scanning bracket by the global visual detection system;

[0052] Performing binarization processing on the first image and the second image respectively, and then performing XOR operation on the binarized first image and the binarized second image, the region with pixel value of 1 in the image after the XOR operation is the region where the ultrasonic probe is located,

[0053] In the first image after binarization, the position with a pixel value of 1 in this region corresponds to the starting position of the ultrasound probe, and in the second image after binarization, the position with a pixel value of 1 corresponds to the ending position of the ultrasound probe. Therefore, the area connecting the starting position and the ending position of the ultrasound probe is the ultrasound probe scanning area.

[0054] The specific steps for identifying the center position of the crawler are as follows:

[0055] Perform an AND operation on the first binarized image and the second binarized image, and calculate the mean of the x and y coordinates of all pixels in the image after the AND operation to obtain the x and y coordinates of the center position of the crawler.

[0056] Step 3: Preset the crawler's crawling trajectory. After each step the crawler takes, the ultrasonic probe scans along a direction perpendicular to the step. After the entire path is scanned, the crawler moves forward one more step. After each step the crawler takes, the local visual inspection system on the crawler takes a picture of the area to be scanned by the ultrasonic probe to obtain a local image of the painted surface. During the movement of the ultrasonic probe, ultrasonic reflection signals are received.

[0057] Step 4: Perform rust identification on the surface image of the local painted structure to obtain the surface rust image identification result; at the same time, perform latent rust identification on the ultrasonic reflection signal to obtain the latent rust identification result; and fuse the surface rust image identification result and the latent rust identification result to obtain the corrosion detection result of the painted structure.

[0058] The specific process for identifying rust on the surface of a partially painted structure is as follows:

[0059] First, a global vision inspection system is used to acquire an overall appearance image of the painted structure to be inspected. The initial paint color of the painted structure is determined by color statistics, and the brightness difference of the paint color at each position under the current illumination is realized. In this embodiment, the purpose of the global vision inspection system in determining the initial paint color of the painted structure to be inspected is to prevent the local vision inspection system from making mistakes.

[0060] In a local visual inspection system, the surface image of a painted structure is acquired. Let the paint color of the current painted structure be RGB(r,g,b). The system acquires RGB images of the entire painted structure surface and performs histogram analysis on each RGB value to calculate the mean RGB value of the paint color. and standard deviation δ, to judge Whether it is within ±k*δ, where k is a constant greater than 0, if yes, then the current paint color is considered normal; if no, then the current paint color is considered abnormal; the rust area in the paint surface is divided by normal and abnormal colors; in this embodiment, k = 3;

[0061] The specific process of the hidden rust identification of the ultrasonic reflection signal is as follows:

[0062] The ultrasonic emission signal includes an interface wave, a bottom wave and noise, and it is judged whether the maximum amplitude of the interface wave is greater than a times the average amplitude of the noise and the maximum amplitude of the bottom wave is lower than b times the average amplitude of the noise, a and b are both constants, a > b, if yes, it is determined that the area detected by the current ultrasonic probe is rusted under the paint layer; if no, it is determined that the area detected by the current ultrasonic probe is not rusted under the paint layer; in this embodiment, a = 2 and b = 1.5, and the specific values of a and b can be determined according to experiments or experience;

[0063] The specific process of fusing the surface rust image recognition result and the hidden rust recognition result is as follows:

[0064] The rust area in the hidden rust recognition result is written into the rust area in the surface rust image recognition result, so as to realize the fusion of the surface rust image recognition result and the hidden rust recognition result.

[0065] In this embodiment, a paint structure maintenance evaluation method using the above paint structure corrosion nondestructive testing method is also involved, which comprises: calculating the rust proportion of the paint structure, judging whether the rust proportion of the paint structure exceeds a preset maintenance setting value, if yes, the paint structure needs to be maintained; if no, the paint structure does not need to be maintained. The preset maintenance setting value in this embodiment is confirmed according to experience, and the rust proportion of the paint structure is the ratio between the rust area on the paint structure and the total area of the paint structure.

[0066] In this embodiment, the paint structure is a storage tank, the global visual detection system is denoted as CCD0, and the local visual detection system is denoted as CCD1, as shown in FIG. 1. Figure 3 As shown in FIG. 2, when the storage tank is a vertical steel storage tank, the crawler adopts a bow-shaped scanning track, the crawler first moves step by step along the circumferential direction, after each step, the scanning support moves along the vertical direction to the step direction with the ultrasonic probe, after the full scanning, the crawler moves one step further, the scanning support moves along the vertical direction to the step direction with the ultrasonic probe, and the process is repeated. When the crawler moves from the starting end to the current circumferential direction track end point, the crawler completes 360-degree turning and moves one step down, and then scans back along the circumferential direction, and the process is repeated until the scanning of the area is completed.

[0067] As shown in FIG. 3, when the storage tank is a horizontal steel storage tank, the crawler adopts a bow-shaped scanning track, the crawler first moves step by step along the circumferential direction, after each step, the scanning support moves along the vertical direction to the step direction with the ultrasonic probe, after the full scanning, the crawler moves one step further, the scanning support moves along the vertical direction to the step direction with the ultrasonic probe, and the process is repeated. When the crawler moves from the starting end to the current circumferential direction track end point, the crawler completes 360-degree turning and moves one step down, and then scans back along the circumferential direction, and the process is repeated until the scanning of the area is completed. Figure 4As shown, when the storage tank is a horizontal steel storage tank, the crawler adopts a bow-shaped scanning track, the crawler first moves step by step along the circumferential direction, after each step, the scanning support with the ultrasonic probe moves along the direction perpendicular to the step direction for scanning, after the full scanning, the crawler moves forward by one step, the scanning support moves for scanning, and the process is repeated. When the crawler moves from the starting end to the current circumferential direction track end point, the crawler completes 360-degree turning, and moves down by one step, and then scans back along the circumferential direction, and the process is repeated until the scanning of the area is completed.

[0068] As shown in the figure, Figure 5 As shown in the figure, the black point represents the center position O1 of the crawler, and the white point represents the center position of the ultrasonic probe. At each black point position, the CCD1 first takes a picture of the area to be scanned by the ultrasonic probe.

[0069] As shown in the figure, Figure 6 As shown in the figure, the surface rust image recognition result and the hidden rust recognition result are fused to obtain the paint structure corrosion detection result.

[0070] The above is only the preferred embodiment of the present application, it should be pointed out that, for the ordinary skilled in the art, without departing from the technical principles of the present application, can make a number of improvements and refinements, these improvements and refinements should also be considered as the protection scope of the present application.

Claims

1. A non-destructive testing method for corrosion of painted structures, characterized in that... It includes a global visual inspection system and an ultrasonic inspection system. The ultrasonic inspection system includes a crawler that can crawl on painted structures, an ultrasonic probe mounted on the crawler, and a local visual inspection system. The non-destructive testing method for corrosion of painted structures includes the following steps: Step 1: Place the crawler on the painted surface to be inspected, and determine the placement of the global vision inspection system so that the field of view of the global vision inspection system can cover the crawler's crawling area. Step 2: Start the global vision inspection system. At this time, the global vision inspection system begins to acquire images within the field of view and identifies and locates the crawler in the image, identifying the center position of the crawler and the scanning area of ​​the ultrasonic probe. Step 3: Preset the crawler's crawling trajectory. After each step the crawler takes, the ultrasonic probe scans along a direction perpendicular to the step. After the entire path is scanned, the crawler moves forward one more step. After each step the crawler takes, the local visual inspection system on the crawler takes a picture of the area to be scanned by the ultrasonic probe to obtain a local image of the painted surface. During the movement of the ultrasonic probe, ultrasonic reflection signals are received. Step 4: Perform rust identification on the surface image of the local painted structure to obtain the surface rust image identification result; at the same time, perform latent rust identification on the ultrasonic reflection signal to obtain the latent rust identification result. The surface corrosion image recognition results and the latent corrosion recognition results are then fused to obtain the corrosion detection results of the painted structure. The specific process for identifying rust in the local painted surface image in step 4 is as follows: First, a global vision inspection system is used to acquire an overall appearance image of the painted structure to be inspected. Then, the initial paint color of the painted structure is determined by color statistics, and the brightness difference of the paint color at each position under the current illumination is realized. In the local visual inspection system, the paint color of the surface of the painted structure is denoted as RGB(r, g, b). By acquiring the RGB image of the entire painted structure surface and performing histogram statistics on each RGB value, the mean and standard deviation δ of the RGB colors are calculated. It is then determined whether the values ​​are within ±k*δ, where k is a constant greater than 0. If so, the current paint color is considered normal; otherwise, it is considered abnormal. The rusted areas in the paint surface are segmented based on the normal and abnormal colors. The specific process for identifying latent corrosion using ultrasonic reflection signals in step 4 is as follows: The ultrasonic transmission signal includes interface wave, bottom wave, and noise. It is determined whether the maximum amplitude of the interface wave is greater than a times the average amplitude of the noise and the maximum amplitude of the bottom wave is less than b times the average amplitude of the noise, where a and b are constants and a > b. If so, it is determined that the area detected by the ultrasonic probe is corroded under the paint layer; otherwise, it is determined that the area detected by the ultrasonic probe is not corroded under the paint layer.

2. The non-destructive testing method for corrosion of painted structures according to claim 1, characterized in that: The crawler is also provided with a scanning bracket, which is arranged perpendicular to the crawler's travel direction. The ultrasonic probe is constrained on the scanning bracket in a manner that allows it to move along the extension direction of the scanning bracket.

3. The non-destructive testing method for corrosion of painted structures according to claim 2, characterized in that: The specific steps for identifying the ultrasound probe scanning area in step 2 are as follows: The crawler is fixed at the starting position of the painted structure to be inspected, while the ultrasonic probe is at the starting point of the scanning bracket. The global vision inspection system is activated, and the ultrasonic probe is moved from the starting point to the ending point of the scanning bracket. The first image when the ultrasonic probe is at the starting point of the scanning bracket and the second image when the ultrasonic probe is at the ending point of the scanning bracket are acquired through the global vision inspection system. The first and second images are binarized separately. Then, an XOR operation is performed on the binarized first and second images. The regions with a pixel value of 1 in the XORed image represent the regions where the ultrasound probe is located. In the first image after binarization, the position with a pixel value of 1 in this region corresponds to the starting position of the ultrasound probe, and in the second image after binarization, the position with a pixel value of 1 corresponds to the ending position of the ultrasound probe. Therefore, the connecting area between the starting position and the ending position of the ultrasound probe is the ultrasound probe scanning area.

4. The non-destructive testing method for corrosion of painted structures according to claim 3, characterized in that: The specific steps for identifying the center position of the crawler are as follows: Perform an AND operation on the first and second binarized images, and then calculate the average of the x and y coordinates of all pixels in the resulting image to obtain the x and y coordinates of the crawler's center position.

5. The non-destructive testing method for corrosion of painted structures according to any one of claims 1 to 4, characterized in that: The specific process of fusing the surface corrosion image recognition results and the latent corrosion recognition results in step 4 is as follows: The rust region in the latent rust identification result is written into the rust region in the surface rust image identification result, thereby achieving the fusion of the surface rust image identification result and the latent rust identification result.

6. A method for evaluating the repair of painted structures using the non-destructive testing method for corrosion of painted structures as described in any one of claims 1 to 5, characterized in that... This includes calculating the corrosion rate of the painted structure, determining whether the corrosion rate exceeds the preset maintenance setting value, and if so, performing maintenance on the painted structure; otherwise, performing maintenance on the painted structure is not required.

Images