A method for evaluating the surface corrosion degree of automobile radiators based on image processing

Abstract

The present invention relates to the field of image data processing technology, and more specifically, to a method for assessing the degree of corrosion on the surface of an automobile radiator based on image processing. The method comprises: acquiring a surface image of the automobile radiator and gray-scaling it to obtain an initial image; performing saliency detection on the initial image using a FT algorithm; and, before the saliency detection, comprehensively evaluating the texture level and horizontal correlation number of each pixel to assess the probability that each pixel is in a support bar region, thereby eliminating pixels belonging to the support bar region before the saliency detection. After the elimination is completed, identifying corrosion regions based on the saliency detection results of the remaining pixels, and assessing the degree of corrosion on the radiator surface based on the size of the corrosion regions. The present invention can improve the accuracy of corrosion region identification, thereby enabling accurate assessment of the degree of corrosion on the radiator surface.

Description

Technical Field

[0001] The present invention relates to the technical field of image data processing, and more particularly to a method for evaluating the degree of corrosion on the surface of an automobile radiator based on image processing. Background Art

[0002] With the global development of the automotive industry and the continuous increase in the number of vehicles on the road, automobile radiators, as core components of the engine cooling system, are exposed to complex environmental conditions such as high temperature, high humidity, and salt spray erosion for a long time. Surface corrosion has become a key risk factor that causes nonlinear attenuation of heat dissipation efficiency, coolant permeability leakage, and local engine overheating failures. Therefore, how to achieve accurate detection of radiator surface corrosion areas has become a technical problem that needs to be solved urgently.

[0003] With the rapid advancement of computer vision and artificial intelligence technologies, image processing technology has become capable of handling complex detection tasks. Among related technologies, the FT (Frequency-Tuned) algorithm for saliency detection based on frequency domain residual analysis can identify abnormal areas by analyzing local contrast features of images.

[0004] However, the radiator surface is designed with equally spaced transverse support strips (duct baffles) to enhance structural stability and optimize aerodynamic characteristics. These strips exhibit periodically arranged high-frequency texture features in the image space domain. As a result, the FT algorithm's residual calculation process incorrectly maps the inherent frequency domain response of such functional structures into local contrast anomaly features, generating false-positive interference signals with similar significance responses to actual corrosion areas. This results in misjudgment of corrosion areas determined based on the significance detection results output by the algorithm, making it impossible to accurately assess the degree of corrosion on the radiator surface. Summary of the Invention

[0005] To address the problem that the presence of support bars on the radiator surface prevents the saliency detection results of radiator surface images based on the FT algorithm from accurately assessing the degree of corrosion on the radiator surface, the present invention provides a method for assessing the degree of corrosion on the surface of an automobile radiator based on image processing. The method comprises:

[0006] Acquire the surface image of the car radiator and perform grayscale processing to obtain the initial image;

[0007] Performing saliency detection on the initial image using the FT algorithm, and removing pixels belonging to the support bar area before saliency detection. After the removal is completed, identifying the corrosion area based on the saliency detection results of the remaining pixels, and evaluating the degree of corrosion on the radiator surface by the size of the corrosion area;

[0008] The method for identifying pixels belonging to the support strip area includes: calculating the texture degree of each pixel based on the grayscale value of each pixel and the consistency of the grayscale values within the neighborhood of each pixel, wherein the texture degree represents the probability that the pixel is in the support strip area;

[0009] Taking each pixel point as the center, traverse horizontally to both sides, count the number of pixels whose grayscale value difference with the central pixel point is less than the first threshold, obtain the horizontal association number of each pixel point, and calculate the high-frequency texture coefficient of each pixel point. The high-frequency texture coefficient is positively correlated with the texture degree and the horizontal association number. The pixel point with a high-frequency texture coefficient greater than the set value is defined as the pixel point of the support strip area.

[0010] The present invention utilizes the horizontal distribution characteristics of pixels in the support bar area and the expression characteristics of grayscale values to evaluate the probability that each pixel belongs to the support bar area, thereby accurately identifying the pixels belonging to the support bar area. Before using the FT algorithm to perform significance detection on the obtained initial image, the pixels belonging to the support bar area are eliminated, which can prevent the FT algorithm from identifying the support bar area as a corrosion area, thereby ensuring the accuracy of corrosion area identification and realizing accurate assessment of the degree of corrosion on the radiator surface.

[0011] Preferably, the high-frequency texture coefficient is also positively correlated with the credibility of each pixel belonging to the support area, and the method for obtaining the credibility includes:

[0012] Taking each pixel as the center, traverse vertically to both sides and define the pixels whose grayscale value difference with the central pixel is less than the second threshold as the same type of pixel;

[0013] The confidence level of each pixel belonging to the support area is calculated, and the confidence level is negatively correlated with the difference between the horizontal association number of the pixel and the horizontal association number of similar pixels.

[0014] By calculating the credibility of each pixel point belonging to the support member area, the present invention can avoid the influence of the presence of dirt on the surface of the radiator on the evaluation of the possibility that each pixel point belongs to the support member area.

[0015] Preferably, the high-frequency texture coefficient satisfies the following relationship:

[0016] ;

[0017] Where, For the High-frequency texture coefficients of pixels; For the The texture level of each pixel; For the The number of horizontal associations of pixels; For the The average level of association of all similar pixels of a pixel; To preset hyperparameters; is the absolute value symbol; is the normalization function.

[0018] The present invention integrates various data to accurately evaluate the possibility that each pixel point belongs to the support area, thereby obtaining an accurate quantitative index, namely the high-frequency texture coefficient, which provides a reference standard for the subsequent identification of pixel points in the support area.

[0019] Preferably, when traversing toward both sides along the vertical direction with each pixel point as the center, if a pixel point appears on either side whose grayscale value difference with the central pixel point is greater than the second threshold, the traversal process on that side is stopped.

[0020] Preferably, the method for obtaining the consistency of the grayscale values within the neighborhood of each pixel point includes:

[0021] The average difference between the grayscale value of each pixel and the grayscale values of the pixels within the four neighborhoods is calculated, and the inverse of the average difference is used as the power of the exponential function to perform power operation to obtain the consistency of the grayscale values within the neighborhood of each pixel.

[0022] The present invention calculates the consistency of the grayscale values within the neighborhood of each pixel point through a negative exponential function, which can effectively suppress the consistency of pixels with large grayscale value differences in the neighborhood.

[0023] Preferably, the texture degree of each pixel is calculated to satisfy the following relationship:

[0024] ;

[0025] Where, For the The texture level of each pixel; For the Gray value of each pixel; is the average grayscale value of all pixels in the initial image; For the The consistency of the grayscale value in the area around the pixel.

[0026] When measuring the grayscale value of each pixel, the present invention provides a reference standard, namely the average grayscale value of all pixels in the initial image, which can effectively eliminate the influence of the overall brightness change of the image on the analysis result.

[0027] Preferably, when traversing horizontally toward both sides with each pixel point as the center, if the grayscale value difference between two consecutive pixels on either side and the central pixel point is greater than the first threshold, the traversal process on that side is stopped.

[0028] Preferably, the corrosion area is identified based on the saliency detection results of the remaining pixels:

[0029] A preset significance threshold is obtained, all pixels whose significance values are greater than the significance threshold are screened, and a connected domain composed of all the screened pixels is used as the erosion area.

[0030] Preferably, the degree of corrosion on the radiator surface is assessed by the size of the corrosion area, including:

[0031] The ratio of the number of pixels in the corrosion area to the number of pixels in the initial image is calculated, and the ratio is used as an evaluation value of the degree of corrosion on the radiator surface.

[0032] Preferably, when grayscale processing is performed on the surface image of the automobile radiator, a weighted average method is used to convert the RGB channel values into grayscale values.

[0033] The present invention has the following effects:

[0034] The present invention calculates the number of horizontal correlations of each pixel point, can divide the boundary of the support area into the support area, and combines the feature of higher grayscale value of the pixel points in the support area to identify the pixel points in the entire support area, so that before using the FT algorithm to perform significance detection on the initial image, all the pixel points belonging to the support area can be eliminated, thereby eliminating the influence of the existence of the support area on the corrosion area identified by the FT algorithm, ensuring the accuracy of the obtained corrosion area, and thus providing a reliable data basis for the evaluation of the degree of corrosion on the radiator surface. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 The present invention is a flowchart of a method for evaluating the surface corrosion degree of an automobile radiator based on image processing according to an embodiment of the present invention. DETAILED DESCRIPTION

[0036] The specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0037] Reference Figure 1 A method for evaluating the surface corrosion degree of an automobile radiator based on image processing includes steps S1 to S3, specifically as follows:

[0038] S1: Obtain a surface image of a car radiator and perform grayscale processing to obtain an initial image.

[0039] Specifically, a high-definition digital camera can be used to capture images of the surface of the car radiator under uniform lighting conditions to ensure that the image is clear and covers the entire surface of the radiator. Then, in order to simplify the subsequent processing process, the captured surface image can be grayscaled.

[0040] In an exemplary embodiment of the present invention, when grayscale processing is performed on the surface image of the automobile radiator, a weighted average method is used to convert RGB channel values into grayscale values.

[0041] Optionally, a maximum value method, an average value method, etc. may be used to convert the RGB channel value of each pixel in the surface image into a grayscale value, thereby converting the collected surface image into a grayscale image to obtain the initial image. This embodiment does not specifically limit the selected grayscale processing method.

[0042] S2: Perform saliency detection on the initial image using the FT algorithm, and remove pixels belonging to the support strip area before the saliency detection.

[0043] It should be noted that there are usually many horizontally arranged support bars between the cooling fins on the surface of the car radiator, also called air duct baffles. Their function is to enhance the structural stability of the radiator and optimize the air circulation path so that the air can pass through the cooling fins more evenly, thereby improving the heat dissipation efficiency.

[0044] However, when using the FT (Frequency-tuned) algorithm to detect corrosion on the surface of a car radiator, the support bar area exhibits significant grayscale differences from surrounding pixels, resulting in high-frequency components and large local contrast differences in the frequency domain. The FT algorithm identifies salient regions by calculating spectral residuals in the frequency domain, and these characteristics of the support bar area can easily be misidentified as salient features, causing the algorithm to mistakenly label them as corrosion areas, resulting in false detections. Therefore, the present invention removes pixels belonging to the support bar area before performing saliency detection using the FT algorithm to avoid false detections.

[0045] Specifically, the following steps can be used to identify the pixels within the support strip area:

[0046] Step 1: Calculate the texture degree of each pixel based on the grayscale value of each pixel and the consistency of the grayscale values within the neighborhood of each pixel. The texture degree represents the probability that the pixel is in the support strip area.

[0047] It should be noted that, through scene analysis, it can be seen that the support bar area is usually made of metal. Due to the high reflectivity of the metal material, the support bar area presents a higher grayscale value in the image; however, there may be noise pixels with abnormally high grayscale values in the initial image, which will interfere with the recognition of the support bar area; since the positions of the noise pixels are randomly distributed, and the support bar area is distributed in a strip-like continuous manner, the grayscale values of its pixels in the local neighborhood are highly consistent.

[0048] Therefore, the present invention can evaluate the texture degree of each pixel point belonging to the support strip area by combining the high grayscale value characteristics of the support strip area and the grayscale consistency characteristics of its neighborhood.

[0049] In an exemplary embodiment of the present invention, the consistency of the grayscale values within the neighborhood of each pixel can be determined by the following steps:

[0050] The average difference between the grayscale value of each pixel and the grayscale values of the pixels within the four neighborhoods is calculated, and the inverse of the average difference is used as the power of the exponential function to perform power operation to obtain the consistency of the grayscale values within the neighborhood of each pixel.

[0051] For example, the The consistency of the gray value in the area around the pixel is recorded as ,but Satisfies the following relationship:

[0052] ;

[0053] Where, For the The average difference between the grayscale value of a pixel and the grayscale values of the pixels in the four neighborhoods. The difference here is the absolute value of the difference between the values (it can also be expressed as the square of the difference between the values); The natural constant An exponential function with base .

[0054] Optionally, the consistency of the grayscale values within the neighborhood of each pixel can also be evaluated based on the variance of the grayscale values within the four neighborhoods or eight neighborhoods of each pixel. It should be noted that the process of determining the four neighborhoods and eight neighborhoods of a pixel is an existing technology in the field of imaging and will not be described in detail in this embodiment.

[0055] Furthermore, after determining the consistency of the grayscale values within the neighborhood of each pixel, the texture degree of each pixel can be calculated. Specifically, the texture degree of each pixel satisfies the following relationship:

[0056] ;

[0057] Where, For the The texture level of each pixel; For the Gray value of each pixel; is the average grayscale value of all pixels in the initial image; For the The consistency of the grayscale value in the area around the pixel.

[0058] in, Reflects the The relative size of the grayscale value of a pixel in the entire initial image. The larger the value, the more likely the pixel is to belong to the support strip area (or possibly a noise pixel), and the greater the texture level of the corresponding pixel. It should be noted that when evaluating the grayscale value of each pixel, the present invention introduces the average grayscale value of the entire image, which provides a global reference standard for grayscale value evaluation, thereby objectively reflecting the grayscale value of each pixel.

[0059] The larger the value, the more consistent the grayscale values in the area around the pixel point, which means that the pixel point is less likely to be a noise pixel point, that is, the pixel point is more likely to belong to the support bar area, and the corresponding pixel point has a relatively large texture degree.

[0060] Step 2: With each pixel as the center, traverse horizontally to both sides, and count the number of pixels whose grayscale value difference with the center pixel is less than a first threshold to obtain the horizontal correlation number of each pixel.

[0061] It should be noted that although the interference of noise pixels has been taken into account when calculating the texture degree of each pixel, pixels located at the edge of the support bar may still be missed. This is because the grayscale values of the pixels at the edge of the support bar are quite different from those of the pixels in the neighborhood, which makes it impossible for these edge pixels to be classified into the support bar area. According to the scene investigation, the support bar area on the surface of the car radiator has a significant horizontal distribution feature. Therefore, the present invention uses this feature to further evaluate the probability of each pixel belonging to the support bar area by analyzing the number of neighboring pixels with small grayscale differences in the horizontal direction.

[0062] For example, Taking a pixel as an example, the process of determining the horizontal correlation number of each pixel is described in detail:

[0063] First Pixel point as the center, along the horizontal direction from the The position of the first pixel is expanded to both sides. When the expanded pixel is aligned with the If the grayscale value difference of pixels (the absolute value of the difference between the values) is less than the first threshold, such as 3, the count is repeated until the preset extension termination condition is reached (such as reaching the image boundary or the grayscale difference of multiple consecutive pixels exceeds the threshold), and the final counting result is used as the first threshold. The number of horizontal associations per pixel.

[0064] In an exemplary embodiment of the present invention, when traversing horizontally to the left and right sides with each pixel point as the center, if the grayscale value difference between two consecutive pixel points and the central pixel point on any side is greater than the first threshold, the traversal process on that side is stopped.

[0065] It should be noted that the larger the number of horizontal associations of any pixel point, the larger the number of pixels with smaller grayscale value differences with the pixel point in the horizontal direction, and further, the more the grayscale value difference between the pixel point and its neighboring pixels in the horizontal direction conforms to the support strip area, and the more likely the pixel point is to belong to the support strip area pixel point.

[0066] Step 3: Calculate the high-frequency texture coefficient of each pixel. The high-frequency texture coefficient is positively correlated with the texture degree and the number of horizontal associations.

[0067] The high-frequency texture coefficient is an indicator used to determine whether a pixel belongs to the support strip region. It should be noted that since the grayscale values of pixels in the support strip region are relatively high, when using the FT algorithm to perform saliency detection on the initial image, the support strip region usually belongs to the high-frequency portion of the image frequency domain. Therefore, this indicator is named the high-frequency texture coefficient here.

[0068] In an exemplary embodiment of the present invention, the high-frequency texture coefficient is also positively correlated with the confidence that each pixel point belongs to the support region. The confidence that each pixel point belongs to the support region can be determined by the following steps:

[0069] (1) Taking each pixel as the center, traverse in the vertical direction to both sides, and define the pixels whose grayscale value difference with the central pixel is less than the second threshold as the same type of pixel points as the corresponding pixel point;

[0070] It should be noted that in actual applications, dust and dirt attached to the radiator surface may significantly reduce the number of pixels with similar grayscale values in the horizontal direction, thereby affecting the accurate identification of the support bar area. To further address this issue, this method introduces vertical analysis. Specifically, if a pixel has multiple similar pixels in the vertical direction (i.e., pixels with similar grayscale values), and the difference in the number of pixels with similar grayscale values in the horizontal direction corresponding to these similar pixels is small, it means that the pixel being analyzed and its similar pixels in the vertical direction are more likely to belong to the same support bar area, which means that the confidence that the pixel being analyzed belongs to the support bar area is greater.

[0071] For example, Taking a pixel as an example, the process of determining similar pixels of each pixel is described in detail:

[0072] First The pixel point is the center, and the vertical direction is from the The position of the first pixel is expanded upward and downward. When the expanded pixel is aligned with the first pixel, If the grayscale value difference of pixels (the absolute value of the difference in values) is less than the second threshold, such as 3, the pixel expanded to is regarded as the first pixel. The expansion ends when the preset extension condition is reached (such as reaching the image boundary or the grayscale difference with the central pixel exceeds the threshold).

[0073] In an exemplary embodiment of the present invention, when traversing vertically to both sides with each pixel point as the center, if a pixel point appears on either side whose grayscale value difference with the central pixel point is greater than the second threshold, the traversal process on that side is stopped.

[0074] (2) Calculate the credibility of each pixel belonging to the support area, and the credibility is negatively correlated with the difference between the horizontal association number of the pixel and the horizontal association number of the same type of pixels.

[0075] For example, Taking a pixel as an example, the calculation of the credibility of each pixel belonging to the support area is described in detail:

[0076] First, determine the The number of horizontal associations of pixels, and the The average value of the horizontal correlation number of all similar pixels of the pixel point; then, calculate the The difference between the horizontal correlation number of the pixel points and the average value (the difference here is the absolute value of the difference between the values); then, the reciprocal of the obtained difference value can be used as the first The confidence level of each pixel belongs to the support area.

[0077] Optionally, the negative correlation between the difference value and the credibility here may be represented in other ways, such as not expressing the negative correlation in the form of a reciprocal but expressing it in the form of a negative natural exponential function.

[0078] Furthermore, after determining the number of horizontal associations, texture degree, and credibility of each pixel belonging to the support area, the high-frequency texture coefficient of each pixel can be calculated. Specifically, the high-frequency texture coefficient of each pixel satisfies the following relationship:

[0079] ;

[0080] Where, For the High-frequency texture coefficients of pixels; For the The texture level of each pixel; For the The number of horizontal associations of pixels; For the The average level of association of all similar pixels of a pixel; To preset hyper parameters, in this embodiment , used to prevent the denominator from being zero; is the absolute value symbol; is the normalization function.

[0081] in, Reflects the The larger the value, the more likely the pixel is to belong to the support area, and the larger the corresponding high-frequency texture coefficient.

[0082] In another embodiment, the relationship: Calculate the high-frequency texture coefficient of each pixel; where, The natural constant An exponential function with base .

[0083] Step 4: Define the pixels whose high-frequency texture coefficient is greater than the set value as the pixels in the support strip area.

[0084] Optionally, the setting value can be set to 0.5. If the high-frequency texture coefficient of any pixel is greater than 0.5, the pixel is determined to be a support area pixel and marked, so that after subsequent significance analysis, such marked pixels are eliminated.

[0085] S3: After the removal is completed, the corrosion area is identified based on the saliency detection results of the remaining pixels, and the corrosion degree of the radiator surface is evaluated by the size of the corrosion area.

[0086] In an exemplary embodiment of the present invention, the identification of the corrosion area can be achieved by the following steps:

[0087] A preset significance threshold is obtained, all pixels whose significance values are greater than the significance threshold are screened, and a connected domain composed of all the screened pixels is used as the erosion area.

[0088] It should be noted that the process of performing saliency detection on an image using the FT algorithm to obtain the normalized saliency value of each pixel is a prior art and will not be described in detail in this embodiment.

[0089] Optionally, the significance threshold can be set to 0.8. If the significance value of any pixel remaining after the elimination operation is completed is greater than 0.8, the location of any pixel is determined to be a corrosion area, so that all pixels with significance values greater than the significance threshold can be screened, and the connected domain composed of all the screened pixels is used as the corrosion area.

[0090] In an exemplary embodiment of the present invention, the evaluation of the corrosion degree of the radiator surface can be achieved by the following steps:

[0091] The ratio of the number of pixels in the corrosion area to the number of pixels in the initial image is calculated, and the ratio is used as an evaluation value of the degree of corrosion on the radiator surface.

[0092] Specifically, the evaluation value satisfies the relationship: Where, is the evaluation value of the corrosion degree of the radiator surface; is the number of pixels in the corrosion area; is the number of pixels in the initial image.

[0093] Optionally, the connectivity of the corrosion area can also be analyzed, for example, the number of connected blocks in the corrosion area and the size of each connected block can be calculated. If the corrosion area is dispersed and has a large number of connected blocks, it may indicate that the corrosion is more serious.

[0094] It should be understood that various alternatives to the embodiments of the invention described herein may be employed in practicing the invention.

Claims

1. A method for evaluating the surface corrosion degree of an automobile radiator based on image processing, characterized in that: include: Acquire the surface image of the car radiator and perform grayscale processing to obtain the initial image; Performing saliency detection on the initial image using the FT algorithm, and removing pixels belonging to the support bar area before saliency detection. After the removal is completed, identifying the corrosion area based on the saliency detection results of the remaining pixels, and evaluating the degree of corrosion on the radiator surface by the size of the corrosion area; The method for identifying pixels belonging to the support strip area includes: calculating the texture degree of each pixel based on the grayscale value of each pixel and the consistency of the grayscale values within the neighborhood of each pixel, wherein the texture degree represents the probability that the pixel is in the support strip area; Taking each pixel point as the center, traverse horizontally to both sides, count the number of pixels whose grayscale value difference with the central pixel point is less than a first threshold, obtain the horizontal association number of each pixel point, calculate the high-frequency texture coefficient of each pixel point, the high-frequency texture coefficient is positively correlated with the texture degree and the horizontal association number, and define the pixel point with a high-frequency texture coefficient greater than a set value as the pixel point of the support strip area; the high-frequency texture coefficient is also positively correlated with the credibility of each pixel point belonging to the support part area. The method for obtaining the credibility includes: Taking each pixel as the center, traverse vertically to both sides and define the pixels whose grayscale value difference with the central pixel is less than the second threshold as the same type of pixel; Calculating the confidence level of each pixel belonging to the support region, wherein the confidence level is negatively correlated with the difference between the horizontal association number of the pixel and the horizontal association number of similar pixels; The high-frequency texture coefficient satisfies the following relationship: ； Where, For the High-frequency texture coefficients of pixels; For the The texture level of each pixel; For the The number of horizontal associations of pixels; For the The average level of association of all similar pixels of a pixel; To preset hyperparameters; is the absolute value symbol; is the normalization function.

2. The method for evaluating the surface corrosion degree of an automobile radiator based on image processing according to claim 1, characterized in that: When traversing toward both sides along the vertical direction with each pixel point as the center, if a pixel point appears on either side whose grayscale value difference with the central pixel point is greater than the second threshold, the traversal process on that side is stopped.

3. The method for evaluating the surface corrosion degree of an automobile radiator based on image processing according to claim 1, characterized in that: The method for obtaining the consistency of the grayscale values within the neighborhood of each pixel point includes: The average difference between the grayscale value of each pixel and the grayscale values of the pixels within the four neighborhoods is calculated, and the inverse of the average difference is used as the power of the exponential function to perform power operation to obtain the consistency of the grayscale values within the neighborhood of each pixel.

4. The method for evaluating the surface corrosion degree of an automobile radiator based on image processing according to claim 3 is characterized in that: The calculation of the texture degree of each pixel satisfies the following relationship: ； Where, For the The texture level of each pixel; For the Gray value of each pixel; is the average grayscale value of all pixels in the initial image; For the The consistency of the grayscale value in the area around the pixel.

5. The method for evaluating the surface corrosion degree of an automobile radiator based on image processing according to claim 1, characterized in that: When traversing horizontally toward both sides with each pixel point as the center, if the grayscale value difference between two consecutive pixels on either side and the central pixel point is greater than the first threshold, the traversal process on that side is stopped.

6. The method for evaluating the surface corrosion degree of an automobile radiator based on image processing according to claim 1, characterized in that: The erosion area is identified based on the saliency detection results of the remaining pixels: A preset significance threshold is obtained, all pixels whose significance values are greater than the significance threshold are screened, and a connected domain composed of all the screened pixels is used as the erosion area.

7. The method for evaluating the surface corrosion degree of an automobile radiator based on image processing according to claim 6, characterized in that: The evaluation of the degree of corrosion on the radiator surface by the size of the corrosion area includes: The ratio of the number of pixels in the corrosion area to the number of pixels in the initial image is calculated, and the ratio is used as an evaluation value of the degree of corrosion on the radiator surface.

8. The method for evaluating the surface corrosion degree of an automobile radiator based on image processing according to claim 1, characterized in that: When graying the surface image of the car radiator, the weighted average method is used to convert the RGB channel values into gray values.

Images