Surface crack intelligent identification method in radiator vibration test

Abstract

The invention relates to the field of image processing, in particular to a surface crack intelligent identification method in a radiator vibration test, and the method comprises the steps: collecting a radiator surface image in real time, and carrying out the preprocessing of the image, and obtaining a radiator gray-scale map; calculating the smoothness of each pixel point of the grey-scale map, the key point density in the neighborhood and the edge degree, and taking the product of the smoothness, the key point density and the edge degree as the optimal degree of the pixel point; obtaining a self-adaptive angular point threshold based on the optimization degree, and carrying out angular point detection on the gray-scale map according to the self-adaptive angular point threshold to obtain all key points through detection; and based on the detected key points, utilizing an LKT optical flow method to realize crack detection on the surface cracks of the radiator. According to the self-adaptive angular point threshold value, the sufficient number of uniformly distributed angular points can be obtained in different areas of the radiator, so that the accuracy and robustness of crack detection are improved.

Description

Technical Field

[0001] The present invention relates to the field of image processing, and more specifically, to an intelligent surface crack recognition method in a radiator vibration test. Background Art

[0002] With the continuous progress of the automotive industry, the performance and complexity of vehicles are increasing. The engine is the "heart" of the vehicle, and its thermal management system is crucial to ensure the normal operation of the engine. The radiator, as the core component of this thermal management system, is responsible for transferring the heat generated by the engine to the coolant, and then dissipating it to the outside through the cooling system, thereby keeping the engine working within a suitable temperature range.

[0003] Since cars are subject to various road conditions and driving conditions during driving, the engine and its accessories (including the radiator) often face a complex vibration environment. This long-term, high-frequency, multi-directional mechanical stress will have a certain impact on the radiator, causing tiny cracks on its surface. Over time, the cracks will gradually expand and eventually lead to coolant leakage. Once the coolant leaks, it will not only affect the heat dissipation efficiency of the radiator, but may also cause the engine to overheat, leading to a series of more serious failures.

[0004] Existing technologies, such as the use of LKT optical flow method, perform crack detection on the surface image of the radiator. When cracks appear on the surface of the radiator, the pixels around the crack area will move relative to each other due to the generation and development of the cracks. By calculating the optical flow of these pixels, the existence and expansion of the cracks can be detected. However, since the structure of the radiator surface is relatively regular and its surface image has a strong regularity, it is difficult for the optical flow method to track and obtain a sufficient number of evenly distributed corner points as key points in the radiator surface image, resulting in poor crack detection results. Summary of the invention

[0005] In order to solve the above-mentioned technical problem that it is difficult to track and obtain a sufficient number of evenly distributed corner points as key points in the radiator surface image, resulting in poor crack detection effect, the present invention provides the following technical solution.

[0006] A method for intelligently identifying surface cracks in a radiator vibration test, comprising: Collect the radiator surface image in real time, and pre-process the image to obtain the radiator grayscale image; Calculate the smoothness of each pixel in the grayscale image, the density of key points in the neighborhood, and the edge degree, and take the product of the smoothness, the density of key points in the neighborhood, and the edge degree as the preference degree of the pixel; obtain the adaptive corner point threshold based on the preference degree, and perform corner point detection on the grayscale image according to the adaptive corner point threshold to detect all key points; Based on the detected key points, the LKT optical flow method is used to realize crack recognition of the radiator surface cracks.

[0007] The present invention determines the importance of a pixel by calculating the degree of smoothness, the density of key points in the neighborhood, and the degree of edge, and uses the product of the three as the degree of preference, which can more comprehensively evaluate the characteristics of the pixel, not only can it screen out the pixels with obvious corner point characteristics, but also can ensure that the distribution of corner points in different areas is relatively uniform to a certain extent. Because the three features of smoothness, key point density, and edge degree describe the characteristics of the pixel and its surrounding area from different angles, taking them into consideration can better capture potential corner points, rather than just relying on a single feature, thereby increasing the possibility of obtaining a sufficient number of corner points; Based on the degree of optimization, an adaptive corner point threshold is obtained. Compared with a fixed threshold, an adaptive threshold can be dynamically adjusted according to the local features of the image, thereby better adapting to the characteristics of images in different regions, avoiding the omission or misjudgment of corner points in certain regions due to improper threshold setting, thereby more accurately detecting corner points and helping to obtain a sufficient number of relatively evenly distributed corner points; According to all the high-quality corner points obtained, the actual deformation and movement of the radiator surface can be better reflected, thereby improving the accuracy and robustness of crack detection.

[0008] Preferably, after obtaining the grayscale image of the heat sink, the method further comprises: According to a preset initial corner point threshold, a corner point detection algorithm is used to process the grayscale image to obtain all initial key points in the grayscale image.

[0009] The preset initial corner point threshold may not be completely applicable to all radiator grayscale images. By obtaining the initial key points and observing their distribution and quantity, the threshold can be verified and adjusted. If there are too many or too few initial key points, or the distribution is unreasonable, the threshold can be increased or decreased accordingly to obtain a more appropriate corner point detection result.

[0010] Preferably, the corner detection algorithm is a Shi-Tomasi corner detection algorithm.

[0011] Preferably, the process of obtaining the smoothness degree includes: Get the gradient value and gradient direction of the pixel point of the grayscale image, calculate the average gradient value in the set neighborhood of the pixel point, normalize the average gradient value using a logarithmic function, and get the smoothness of the pixel point.

[0012] The smoothness information can help determine the distribution of corner points in different areas. In relatively smooth areas, there may be fewer corner points, while in areas with rich textures and complex edges, the corner points may be denser.

[0013] Preferably, the process of obtaining the density of key points in the neighborhood includes: Count the number of pixels that have been detected as initial key points in the neighborhood of the pixel, calculate the ratio of the number of pixels that have been detected as initial key points in the neighborhood of the pixel to the total number of neighboring pixels of the pixel and then normalize it to obtain the density of key points in the neighborhood of the pixel.

[0014] In the corner detection algorithm, there may be cases where some misdetected or noise points are regarded as corners. By analyzing the density of key points, the corners located in the high-density area can be regarded as more reliable corners because they are supported by more similar key points around them, while the isolated or low-density area corners may be noise or misdetected points.

[0015] Preferably, the process of obtaining the edge degree includes: Based on the gradient direction, filter out the angle-similar pixels of the pixel, calculate the absolute difference between the gradient value of the pixel and the gradient values of each angle-similar pixel, and perform normalization processing on the absolute difference. Then, perform cumulative summation on all the normalized neighboring pixels with similar angles, and perform average processing on the cumulative result to obtain the average similarity degree of the pixel and its neighboring pixels with similar angles in terms of gradient magnitude, which is used as the edge degree of the pixel.

[0016] In the corner detection algorithm, it is usually desired to find the points that are both in the significantly edge area and have unique features as corners. By calculating the edge degree of the pixel, the areas with a higher probability of being on the edge can be pre-screened, and then corners can be further searched and determined within these areas. This can greatly reduce the range of corner search and improve the efficiency and accuracy of corner extraction.

[0017] Preferably, the adaptive corner threshold satisfies the relational expression: ; where is the adaptive corner threshold corresponding to the pixel, is the preference degree of the pixel, is the initial corner threshold.

[0018] By introducing the preference degree of the pixel, the corresponding corner threshold can be adjusted according to the characteristics of each pixel. For those pixels that are more likely to be corners (i.e., pixels with a higher preference degree), their adaptive corner thresholds will be relatively smaller, which makes these pixels more likely to meet the threshold conditions during the corner judgment process and thus be accurately detected as corners, reducing the misjudgment or missed judgment cases that may be caused by a unified threshold.

[0019] Preferably, the corner detection of the grayscale image according to the adaptive corner threshold includes: Calculate the covariance matrix of the pixel values within the neighborhood of each pixel in the grayscale image. For the covariance matrix of each pixel, perform eigenvalue decomposition to obtain the eigenvalues and corresponding eigenvectors. Denote the ratio of the minimum eigenvalue of the pixel to the adaptive corner threshold as the corner ratio. Screen the pixels whose corner ratio is greater than a preset threshold, and take the pixel with the largest corner ratio as the key point. Then recalculate the adaptive corner threshold for other pixels within the neighborhood, and gradually find all the key points.

[0020] Preferably, the process of obtaining the smoothness includes: Take the reciprocal of the maximum gradient magnitude among the neighborhood pixels of the pixel as the smoothness of the pixel.

[0021] Preferably, the process of obtaining the key point density within the neighborhood includes: Divide the preset neighborhood of the pixel into multiple sub - neighborhoods. For each sub - neighborhood, count the number of pixels that have been detected as initial key points and the total number of pixels within the sub - neighborhood. Take the ratio of the number of pixels that have been detected as initial key points to the total number of pixels within the sub - neighborhood as the key point density within the sub - neighborhood. Calculate the average value of the key point densities within all sub - neighborhoods to obtain the key point density within the neighborhood of the pixel.

[0022] The beneficial effects of the present invention are: By calculating the smoothness, key point density within the neighborhood, and edge degree of each pixel in the grayscale image, and taking the product of the three as the preference degree of the pixel, a variety of characteristic information of the pixels in the image is comprehensively considered. This multi - dimensional feature fusion method can more comprehensively and accurately reflect the characteristics that the pixel may belong to the crack area, avoiding misjudgment caused by a single feature. Then, based on the preference degree of the pixel, dynamically adjust the threshold of corner detection, so that corner detection can better adapt to the image characteristics of different regions, avoiding false detection or missed detection caused by a fixed threshold, and further improving the accuracy of crack detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 is the flowchart of the method from step S1 to step S3 in an intelligent surface crack recognition method for a radiator vibration test in an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0024] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention.

[0025] Refer to Figure 1 , an intelligent surface crack recognition method for a radiator vibration test includes steps S1 - S3, specifically as follows: S1: Collect the image of the radiator surface, and preprocess the image to obtain the grayscale image of the radiator.

[0026] In one embodiment, place the radiator in a suitable position, and use a high-resolution industrial camera to capture the image of the radiator surface in real time. Since the captured image may be affected by environmental noise during the shooting process, it is necessary to perform denoising processing on the captured surface image (such as using methods like mean filtering and Gaussian filtering to remove noise).

[0027] Further convert the above-captured image of the radiator surface into a grayscale image.

[0028] After obtaining the grayscale image, set a suitable initial corner threshold according to experience or prior experimental results. In the embodiment of the present invention, it is initially set to 0.04. Then, use a corner detection algorithm (such as Shi-Tomasi corner detection) to process the above-captured grayscale image to obtain the initial corners in the grayscale image, mark them as initial key points, and record the position information (such as row number and column number) of all pixel points determined to be initial key points. The corner detection algorithm is a prior art and will not be elaborated here.

[0029] S2: Calculate the smoothness degree, the density of key points in the neighborhood, and the edge degree of each pixel point in the grayscale image, and take the product of the smoothness degree, the density of key points in the neighborhood, and the edge degree as the preference degree of the pixel point. Obtain an adaptive corner threshold based on the preference degree, and perform corner detection on the grayscale image according to the adaptive corner threshold to detect all key points.

[0030] Although the initial key points in the grayscale image are obtained in the above S1, the initial corner threshold is set based on experience or preset values, and it may not be able to adapt to the characteristics of all regions in the radiator surface image. Especially in the radiator surface image, there are regions with different features such as smooth regions, texture regions, and edges. Using a fixed initial corner threshold may result in insufficient corner points in some regions (such as smooth regions), missing important feature points, while there are too many corner points in other regions (such as regions with complex textures), causing information redundancy and difficulties in subsequent processing.

[0031] In a smooth region, the gradient change of pixel points is small. If a fixed corner threshold is used, it may be difficult to detect corner points. At the same time, in order to avoid over-concentration of corner points, it is also necessary to consider the number of key points that already exist in the neighborhood of each pixel point.

[0032] In one embodiment, first, for each pixel point in the grayscale image, use four directions respectively ( , , , Perform a convolution operation using the Sobel operator in ( ), obtain the gradient values of each pixel point in four directions, further compare the magnitudes of the gradient values in these four directions, take the largest gradient value as the gradient value of the pixel point, and take the direction corresponding to the largest gradient value as the gradient direction of the pixel point.

[0033] Next, for each pixel point in the grayscale image, obtain all the pixel points within its 9×9 neighborhood, mark these pixel points as neighborhood pixel points, calculate the average gradient value within the pixel point neighborhood, and then normalize this average gradient value using the logarithmic function to obtain the smoothness of the pixel point. The smoothness reflects the texture change situation in the area where the pixel point is located. In a smooth area, the gradient values of pixel points change relatively gently, while corner points are usually located in areas with large gradient changes in the image. By calculating the smoothness, smooth areas and areas that may contain corner points can be distinguished.

[0034] Exemplarily, the smoothness of the above pixel point is expressed by the relational expression:

[0035] In the formula, is the smoothness of the pixel point, is the average gradient value within the pixel point neighborhood.

[0036] When is relatively small, the value of is also relatively small, then

[0037] the value of

[0038] is close to 1, indicating that the neighborhood of this pixel point is relatively smooth; conversely, the gradient change in the neighborhood of this pixel point is complex.

[0039] In the formula, is the key point density within the pixel point neighborhood, is the number of pixel points within the pixel point neighborhood that have been detected as initial key points, is the total number of neighborhood pixel points of the pixel point.

[0040] It should be noted that the edge pixel points in the grayscale image have special properties, and the gradient values of adjacent pixel points are similar. If a fixed threshold is directly used for corner detection, these edge pixel points may be misselected as corners, resulting in errors in the optical flow method detection.

[0041] Therefore, by considering the degree of edge, the corner detection threshold can be appropriately adjusted in the edge area to avoid selecting edge pixel points. Furthermore, by comprehensively considering the smoothness degree, key point density, and edge degree of pixel points, the preference degree of each pixel point is calculated, and the corner threshold is adaptively adjusted according to the preference degree. In this way, pixel points with a high preference degree (i.e., pixel points in the smooth area, with a low key point density and in a non-edge area) have a lower corner threshold and are more likely to be detected as corners; while pixel points with a low preference degree have a higher corner threshold, reducing the possibility of false detection.

[0042] In another embodiment, compared with using the average value of the gradients of all neighborhood pixel points, using the maximum value for simplified calculation, specifically, the reciprocal of the maximum value of the gradient magnitudes among the neighborhood pixel points of the pixel point is used as the smoothness degree of the pixel point, that is, the relational expression is satisfied as:

[0043] In the formula, is the smoothness degree of the pixel point, is the maximum value of the gradient magnitudes among the neighborhood pixel points of the pixel point, is the total number of neighborhood pixel points of the pixel point. Among them, 1 is added to the denominator of the formula to avoid the denominator being 0, and adding 1 to the denominator can smooth the maximum gradient value and avoid the small change in the gradient value having too much influence on the calculation of the smoothness degree.

[0044] In another embodiment, considering the distribution of key points in the neighborhood, the neighborhood of the pixel point is further divided into multiple sub-neighborhoods. For example, a 9×9 neighborhood can be divided into 9 3×3 sub-neighborhoods.

[0045] For each sub-neighborhood, count the number of pixel points that have been detected as initial key points and the total number of pixel points in the sub-neighborhood. Take the ratio of the number of pixel points that have been detected as initial key points to the total number of pixel points in the sub-neighborhood as the key point density in the sub-neighborhood, and average the key point densities in all sub-neighborhoods to obtain the key point density in the neighborhood of the pixel point.

[0046] In one embodiment, for each pixel point in the grayscale image, filter out the neighborhood pixel points whose difference from its gradient angle is not (in this embodiment, by selecting the difference in gradient directions to be Pixels that can better capture features orthogonal to the gradient direction of the current pixel, thereby improving the accuracy of corner detection) are denoted as angle-similar pixels. For example, if the gradient direction of this pixel is , then pixels with gradient directions within can be regarded as angle-similar pixels.

[0047] Then calculate the absolute difference between the gradient value of this pixel and the gradient values of each angle-similar pixel, and normalize this absolute difference. Further, accumulate the normalized values of all angle-similar neighborhood pixels, and average the accumulated result to obtain the average similarity degree of this pixel and its angle-similar neighborhood pixels in terms of gradient magnitude, which is used as the edge degree of this pixel. The edge degree focuses on whether the pixel is located at the edge of the image. Pixels at the edge often have larger gradient values, but these points are not necessarily corners. By considering the edge degree, pixels located at the edge but not corners can be further filtered out, reducing the false detection rate.

[0048] According to the above operations, the smoothness, key point density within the neighborhood, and edge degree of all pixels in the grayscale image can be obtained, and the product of these three factors is used as the preference degree of the pixel.

[0049] The image features in different regions vary greatly, and using a fixed corner threshold may not accurately detect all corners. By calculating the preference degree of each pixel and adjusting the corner threshold according to the preference degree, corner detection can be made more adaptive to the local features of the image. For pixels with a high preference degree, the possibility of them being corners is relatively large, so the corner threshold can be appropriately reduced to improve the sensitivity of corner detection; while for pixels with a low preference degree, the corner threshold can be increased to reduce the false detection rate.

[0050] Exemplarily, the expression for obtaining the adaptive corner threshold based on the preference degree is as follows:

[0051] In the formula, is the adaptive corner threshold corresponding to the pixel, is the preference degree of the pixel, is the initial corner threshold.

[0052] Furthermore, corner detection is performed on the grayscale image according to the adaptive corner threshold, and all key points are detected. The specific process is as follows: Calculate the covariance matrix of all pixels in the image, perform eigenvalue decomposition on the covariance matrix of each pixel to obtain eigenvalues and corresponding eigenvectors, and select the minimum eigenvalue; denote the ratio of the minimum eigenvalue to the adaptive corner threshold corresponding to the pixel calculated above as the corner ratio. If this ratio is greater than 1, it indicates that the pixel may be a corner point; among all the pixels with a corner ratio greater than 1, select the pixel with the largest ratio as the key point; for the newly selected key point, recalculate the adaptive corner threshold of other pixels within its neighborhood range; continue to find new key points until the corner ratio of all pixels is less than 1; the final set of corner points obtained is all the required key points.

[0053] S3: Based on the detected key points, use the LKT optical flow method to achieve crack recognition of the cracks on the radiator surface.

[0054] Under normal circumstances, the movement trajectories of the key points on the radiator surface should be relatively regular and consistent during periodic vibration. However, if there are cracks on the radiator surface, the movement characteristics of the area near the cracks will change due to the influence of the cracks, resulting in abnormal movement trajectories of the key points. By analyzing the movement trajectories of the key points, detect whether there is local movement abnormality caused by cracks, so as to judge whether there are cracks on the radiator surface.

[0055] According to all the key points obtained in the above S2, use the LKT optical flow method to track the movement trajectories of these key points during the periodic vibration of the radiator. If there are cracks on the radiator surface, during the vibration process, the movement states of the materials near the cracks will be different from those in the normal area. For example, the cracks may cause elastic deformation, discontinuous displacement, etc. of the local materials. By analyzing the movement trajectories of the key points, local movement abnormalities caused by cracks can be found. For example, the points near the cracks may show different movement directions, speeds or accelerations from the surrounding points.

[0056] Based on the above detected local movement abnormalities, it can be inferred whether there are cracks on the radiator surface. If the movement trajectories of the key points in a certain area deviate significantly from the normal range, or there are discontinuous movement changes, it is considered that there are cracks in this area, and the position and range of the cracks are roughly determined, providing a basis for subsequent technicians to carry out repairs.

[0057] It should be noted that for those of ordinary skill in the art, without departing from the concept of the present invention, several deformations and improvements can still be made, and these all belong to the protection scope of the present invention. Therefore, the protection scope of this invention patent shall be subject to the appended claims.

Claims

1. An intelligent recognition method for surface cracks in the vibration test of a radiator, characterized in that, Including: Collecting the surface image of the radiator in real time and preprocessing the image to obtain the grayscale image of the radiator; Calculating the smoothness, the density of key points in the neighborhood, and the edge degree of each pixel point in the grayscale image, and taking the product of the smoothness, the density of key points in the neighborhood, and the edge degree as the preference degree of the pixel point; obtaining an adaptive corner threshold based on the preference degree, and performing corner detection on the grayscale image according to the adaptive corner threshold to detect all key points; Based on the detected key points, using the LKT optical flow method to realize crack identification of the cracks on the radiator surface.

2. The intelligent identification method for surface cracks in the vibration test of a radiator according to claim 1, characterized in that, After obtaining the grayscale image of the radiator, it further includes: Processing the grayscale image using a corner detection algorithm according to a preset initial corner threshold to obtain all initial key points in the grayscale image.

3. The intelligent identification method for surface cracks in the vibration test of a radiator according to claim 2, wherein The corner detection algorithm is the Shi-Tomasi corner detection algorithm.

4. The intelligent identification method for surface cracks in the vibration test of a radiator according to claim 3, characterized in that, The process of obtaining the smoothness includes: Obtaining the gradient value and gradient direction of the pixel point in the grayscale image, calculating the average gradient value within the set neighborhood of the pixel point, and normalizing the average gradient value using a logarithmic function to obtain the smoothness of the pixel point.

5. The intelligent recognition method for surface cracks in the vibration test of a radiator according to claim 4, characterized in that, The process of obtaining the density of key points in the neighborhood includes: Counting the number of pixel points that have been detected as initial key points within the neighborhood of the pixel point, calculating the ratio of the number of pixel points that have been detected as initial key points within the neighborhood of the pixel point to the total number of neighborhood pixel points of the pixel point and then normalizing it to obtain the density of key points in the neighborhood of the pixel point.

6. The intelligent recognition method for surface cracks in the vibration test of a radiator according to claim 5, characterized in that, The process of obtaining the edge degree includes: Based on the gradient direction, screening out the angle-similar pixel points of the pixel point, calculating the absolute difference between the gradient value of the pixel point and the gradient values of each angle-similar pixel point, and performing normalization processing on the absolute difference. Then, performing cumulative addition on all normalized neighborhood pixel points with similar angles, and performing average processing on the cumulative result to obtain the average similarity degree of the pixel point and its neighborhood pixel points with similar angles in terms of gradient amplitude, which is used as the edge degree of the pixel point.

7. An intelligent identification method for surface cracks in the vibration test of a radiator according to claim 6, characterized in that, The relational expression satisfied by the adaptive corner threshold is: ; where, is the adaptive corner threshold corresponding to the pixel point, is the preference degree of the pixel point, is the initial corner threshold.

8. The intelligent identification method for surface cracks in the vibration test of a radiator according to claim 7, characterized in that The performing corner detection on the grayscale image according to the adaptive corner threshold includes: Calculating the covariance matrix of the pixel values within the neighborhood range of each pixel point in the grayscale image. For the covariance matrix of each pixel point, performing eigenvalue decomposition to obtain eigenvalues and corresponding eigenvectors; recording the ratio of the minimum eigenvalue of the pixel point to the adaptive corner threshold as the corner ratio; screening out the pixel points with a corner ratio greater than a preset threshold, and taking the pixel point with the largest corner ratio as the key point, and recalculating the adaptive corner threshold of other pixel points within the neighborhood range, and gradually finding all key points.

9. The intelligent recognition method for surface cracks in the vibration test of a radiator according to claim 3, characterized in that The process of obtaining the smoothness includes: Taking the reciprocal of the maximum value of the gradient amplitude among the neighborhood pixel points of the pixel point as the smoothness of the pixel point.

10. The intelligent identification method for surface cracks in the vibration test of a radiator according to claim 1, characterized in that, The process of obtaining the density of key points in the neighborhood includes: Divide the preset neighborhood of a pixel into multiple sub - neighborhoods. For each sub - neighborhood, count the number of pixels that have been detected as initial key points and the total number of pixels in the sub - neighborhood. Take the ratio of the number of pixels that have been detected as initial key points to the total number of pixels in the sub - neighborhood as the key - point density in the sub - neighborhood. Average the key - point densities in all sub - neighborhoods to obtain the key - point density within the pixel neighborhood.

Images