Complex vehicle body surface coating defect detection method based on 2.5 D phase deflection technology

Abstract

The invention provides a 2.5 D phase deflection technology-based complex vehicle body surface coating defect detection method. The method comprises the following steps of: 1, projecting four cosine phase diagrams along an X direction and four cosine phase diagrams along a Y direction to the surface of a vehicle body coating through a programmable light source, and capturing eight cosine phase diagrams reflected by the surface of the vehicle body coating by using a high-precision industrial camera; 2, carrying out fusion processing on eight cosine phase diagrams captured by a camera by adopting a phase deflection technology; and step 3, gradient information of the wrapped phase diagram is extracted and fused, and a shape diagram is generated. The image generated by the technology can detect automobile body coating defects which are difficult to identify by a traditional method, can more clearly reflect fine defects such as sagging, scratches, pits and protrusions on the surface of the automobile coating, and also shows excellent detection capability on the complicated surface of the automobile body coating. According to the technology, the defect detection capability is remarkably improved, the subsequent defect extraction process is simplified, and the detection efficiency and accuracy are greatly improved.

Description

Technical Field

[0001] The present invention relates to the field of image enhancement and industrial defect detection, and in particular to a method for detecting coating defects on a complex vehicle body surface based on 2.5D phase deflectometry. Background Art

[0002] As the demand for high-precision surface measurement in modern industry and manufacturing continues to increase, traditional optical technology has gradually become insufficient in meeting these high requirements, especially when dealing with automotive coating inspection with complex body surfaces. This challenge is particularly prominent. The surface of automotive coatings is usually highly glossy or highly reflective, and traditional industrial cameras and light source technologies often have difficulty accurately capturing tiny surface defects such as pits and scratches. Even if defect images can be captured, these images often lack sufficient details and clarity, making subsequent defect analysis and processing extremely difficult. Especially in areas with complex geometries and uneven light reflections such as the rear of the car, traditional inspection methods face greater challenges and are prone to missed or false inspections. This limitation forces companies to invest a lot of manpower and material resources in manual inspection and analysis, significantly increasing production costs and inspection time. In the automotive manufacturing process, accurate detection of body coating defects is crucial to product quality control. Summary of the invention

[0003] The purpose of the present invention is to provide a method for detecting coating defects on complex vehicle body surfaces based on 2.5D phase deflectometry, which is particularly suitable for the case where the surface of a vehicle body has complex curves in some areas; for example, in areas with complex geometric shapes and uneven light reflection, such as the rear of a vehicle, traditional detection methods face greater challenges and are prone to missed detection or false detection. This limitation forces companies to invest a lot of manpower and material resources in manual detection and analysis, significantly increasing production costs and detection time.

[0004] The technical solution adopted by the present invention is as follows: A method for detecting coating defects on a complex vehicle body surface based on 2.5D phase deflectometry comprises the following steps:

[0005] Step 1: Project 4 cosine phase images along the X direction and 4 cosine phase images along the Y direction onto the coating surface of the vehicle body through a programmable light source and use an industrial camera to capture 8 cosine phase images reflected from the coating surface of the vehicle body;

[0006] Step 2: Use phase deflectometry to fuse the eight cosine phase images captured by the camera to generate a standard image, a specular reflection image, a diffuse reflection image, a wrapping phase image in the X direction, and a wrapping phase image in the Y direction;

[0007] Step 3: Extract the gradient information of the wrapped phase image, fuse the obtained gradient information, and generate a shape map.

[0008] Furthermore, in step 1, the programmable light source projects 4 cosine phase images along the X direction and 4 cosine phase images along the Y direction respectively, and the phase change between adjacent images is a quarter cycle.

[0009] Furthermore, in step 1, the eight cosine phase images captured by the industrial camera are all grayscale images.

[0010] Furthermore, in step 3, when generating the shape map, the gradient information of the wrapped phase map in the X and Y directions is extracted by applying the gradient operator, and then the obtained gradient values ​​are offset and modulo operations are performed to process the phase wrapping effect, and finally, the processed results are fused to generate the shape map.

[0011] Furthermore, in step 2, phase deflectometry is used to fuse the eight original cosine phase images to generate a standard image, a specular reflection image, a diffuse reflection image, a wrapping phase image in the X direction, and a wrapping phase image in the Y direction. The specific formula is as follows:

[0012]

[0013] Among them, I(x,y) represents the pixel gray value, A(x,y) represents the background light intensity, B(x,y) represents the modulated light intensity, represents phase and δ represents phase shift.

[0014] Furthermore, the standard image is generated by averaging and fusing eight original cosine phase images captured by the industrial camera.

[0015] Furthermore, a specular reflection map is generated by extracting the specular reflection components in the X direction and the Y direction.

[0016] Furthermore, the diffuse reflection map is generated by fusing the standard reflection map and the specular reflection map.

[0017] Furthermore, by solving the phase Generate a wrapped phase map in the X direction and a wrapped phase map in the Y direction.

[0018] Furthermore, the diffuse reflection image is used to identify foreign matter and dirt on the surface using the following steps:

[0019] S2.1. Image preprocessing: Use a Gaussian kernel to perform a convolution operation on the diffuse reflectance image to reduce random noise in the image and enhance the image contrast. Finally, the image grayscale value is normalized to 0 to 255 to obtain the processed image.

[0020] S2.2, anomaly detection: calculating the gray value and standard deviation of the processed image and setting a dynamic detection threshold; using a sliding window to analyze local features and mark abnormal areas that exceed the threshold range;

[0021] S2.3, feature extraction: perform connected domain analysis on the marked abnormal area; extract morphological features of the abnormal area; calculate texture features of the abnormal area;

[0022] S2.4, Classification analysis: Classify abnormal areas into foreign matter or dirt.

[0023] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0024] 1. The present invention provides a complex vehicle body surface coating defect detection method based on 2.5D phase deflectometry, which is used for automobile coating surface detection. The technology generates different types of images to detect different types of defects. The standard image provides the overall view of the coating surface and the distribution of defects; the mirror reflection image is used to detect scratches on the vehicle body surface; the diffuse reflection image is used to identify foreign matter and dirt; the shape image is used to detect scratches, bumps, sagging and other surface unevenness defects;

[0025] 2. A complex body surface coating defect detection method based on 2.5D phase deflectometry can detect small bumps and scratches on the body coating that are difficult to identify with traditional methods by generating shape maps. This technology relies on the high-precision measurement capability of phase deflectometry to accurately capture the tiny shape changes on the body coating surface, thereby accurately identifying subtle surface defects and greatly improving the detection capability of complex body surface defects;

[0026] 3. The complex body surface coating defect detection method based on 2.5D phase deflectometry effectively eliminates the interference of the body background through the precise measurement of phase deflectometry, ensuring the accuracy and reliability of the detection of automobile coating surface defects. This technology is also suitable for dealing with complex body surfaces, such as the rear of the car and other areas with complex geometric shapes and uneven light reflection. In these areas that are difficult to accurately identify with traditional detection methods, it can still effectively capture and identify tiny pits, scratches and other defects, thereby improving the detection accuracy of complex body surfaces. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 is a flow chart of the present invention;

[0028] Figure 2 A cosine phase diagram projected by the programmable light source of the present invention;

[0029] Figure 3 The industrial camera in the present invention captures the cosine phase image reflected by the coating surface of the vehicle body;

[0030] Figure 4 It is the standard diagram in the present invention;

[0031] Figure 5 is the specular reflection diagram in the present invention;

[0032] Figure 6 is the diffuse reflection image in the present invention;

[0033] Figure 7 It is the wrapping phase diagram in the X direction and the Y direction in the present invention;

[0034] Figure 8 is the shape diagram in the present invention;

[0035] Fig. 9 A local detail diagram of the shape diagram in the present invention. DETAILED DESCRIPTION

[0036] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0037] In the description of the present invention, it should be noted that the orientations or positional relationships indicated by the terms "center", "up", "down", "left", "right", "vertical", "horizontal", "inside", "outside", etc. are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance. Among them, the terms "first position" and "second position" are two different positions, and the first feature "above", "above" and "above" the second feature include the first feature being directly above and obliquely above the second feature, or simply indicates that the first feature is higher in level than the second feature. The first feature "below", "below" and "below" the second feature include the first feature being directly below and obliquely below the second feature, or simply indicates that the first feature is lower in level than the second feature.

[0038] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0039] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be understood as limiting the present invention.

[0040] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0041] Example 1

[0042] Figure 1 As shown: A method for detecting coating defects on complex vehicle body surfaces based on 2.5D phase deflectometry includes the following steps:

[0043] Step 1: Use a programmable light source to project 4 cosine phase images along the X direction and 4 cosine phase images along the Y direction onto the body coating surface, and use a high-precision industrial camera to capture 8 cosine phase images reflected by the body coating surface. During the projection process, the industrial camera synchronously captures a cosine phase image each time the display projects, and the cosine phase images are all grayscale images. The phase of each of the 4 cosine phase images along the X direction and the 4 cosine phase images along the Y direction increases by a quarter cycle relative to the previous image, thereby ensuring that sufficient information can be obtained in each direction to calculate the complete phase image, thereby performing accurate surface shape analysis and defect detection.

[0044] Step 2: Use phase deflectometry to fuse the eight cosine phase images captured by the camera to generate a standard image, a specular reflection image, a diffuse reflection image, a wrapping phase image in the X direction, and a wrapping phase image in the Y direction;

[0045] The specific formula is as follows:

[0046]

[0047] Among them, I(x,y) represents the pixel gray value, A(x,y) represents the background light intensity, B(x,y) represents the modulated light intensity, represents phase and δ represents phase shift.

[0048] The above formula describes the light intensity distribution of each captured raw cosine phase image, where:

[0049] The standard image is generated by averaging and fusing the eight original cosine phase images captured by the industrial camera. The standard image is generated by averaging the eight cosine phase images to eliminate the cosine modulation term and retain the background light intensity. When generating the standard image, since the cosine modulation term is eliminated, only the background light intensity A(x,y) is retained. A(x,y) directly reflects the surface characteristics, thus providing the overall picture of the coating surface and the defect distribution.

[0050] The specular reflection map is generated by fusing the modulated light intensity of 8 original cosine phase images. The specular reflection map extracts the specular reflection component of the surface and reflects the smoothness of the surface. Therefore, it can accurately reflect the microscopic morphology changes of the surface. The surface roughness changes caused by scratches will significantly affect the specular reflection characteristics, forming a clear contrast in the specular reflection map.

[0051] The diffuse reflection map is generated by fusing the standard map and the specular reflection map. The diffuse reflection map is obtained by removing the specular reflection component and reflects the diffuse reflection characteristics of the surface. Since foreign matter and dirt usually cause changes in local light intensity, which appear as abnormal brightness or color changes in the diffuse reflection map, foreign matter and dirt on the surface can be identified by analyzing these anomalies in the diffuse reflection map.

[0052] By solving the phase, the X-direction wrapping phase map and the Y-direction wrapping phase map are generated. The X-direction and Y-direction wrapping phase maps can reveal the local changes in surface height. Scratches and pits can cause phase mutations or discontinuities, which are very obvious in the phase map.

[0053] By fusing the eight cosine phase images, various characteristic information of the surface to be tested can be extracted, providing basic data for subsequent analysis and processing.

[0054] Step 3: Extract the gradient information of the wrapped phase image, fuse the obtained gradient information, and generate a shape map.

[0055] The wrapped phase images in the X and Y directions are processed to extract phase gradient information. The gradient information reflects the rate of change of the phase in space and can reveal the local morphological characteristics of the surface. Due to the periodic characteristics of the phase information, the gradient information may have a wrapping effect. The present invention eliminates the phase wrapping effect by offsetting and modulo operations on the gradient information, ensures the continuity and accuracy of the gradient information, and fuses the processed gradient information in the X and Y directions to generate a surface shape map, which provides height distribution information on the surface of the object, thereby being used to detect surface defects of the automobile body.

[0056] Example 2

[0057] The difference between this embodiment and embodiment 1 is that in step 2, by analyzing these anomalies in the diffuse reflection image to identify foreign matter and dirt on the surface, the following steps are included:

[0058] S2.1. Image preprocessing: Use a Gaussian kernel to perform a convolution operation on the diffuse reflectance image to reduce random noise in the image and enhance the image contrast. Finally, the image grayscale value is normalized to 0 to 255 to obtain the processed image.

[0059] S2.2, anomaly detection: calculating the gray value and standard deviation of the processed image and setting a dynamic detection threshold; using a sliding window to analyze local features and mark abnormal areas that exceed the threshold range;

[0060] Traversing each pixel in the processed image, accumulating the grayscale values ​​of all pixels, and dividing the sum by the total number of pixels to obtain an average grayscale value;

[0061] Traversing each pixel of the processed image, calculating the square of the difference between its grayscale value and the average grayscale value, accumulating the square differences and dividing the sum of the squares by the total number of pixels to obtain the variance, taking the square root of the variance to obtain the standard deviation;

[0062] The dynamic detection threshold is the mean gray value ± 2 times the standard deviation;

[0063] The processed image is traversed to detect whether the grayscale value of the local area exceeds the dynamic detection threshold range. If exceeded, the pixels exceeding the range are marked as abnormal.

[0064] S2.3, feature extraction: perform connected domain analysis on the marked abnormal area; extract morphological features of the abnormal area; calculate texture features of the abnormal area;

[0065] Identify and extract connected abnormal regions in images;

[0066] Calculate the texture features of each abnormal region, including area, perimeter, and shape factor;

[0067] Gray-level co-occurrence matrix features, including contrast and homogeneity, were calculated for each region.

[0068] S2.4, Classification analysis: Classify abnormal areas into foreign matter or dirt;

[0069] When classifying anomaly areas:

[0070] Those with high contrast, nearly regular shape, large area and low homogeneity are foreign objects;

[0071] Dirt is characterized by low contrast, irregular shape, small area, and high homogeneity.

[0072] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. For those skilled in the art, various obvious changes, readjustments and substitutions can be made without departing from the protection scope of the present invention. It is not necessary and impossible to list all the embodiments here. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the claims of the present invention.

Claims

1. A method for detecting coating defects on complex vehicle body surfaces based on 2.5D phase deflectometry, characterized in that: The following steps are involved: Step 1: Project 4 cosine phase images along the X direction and 4 cosine phase images along the Y direction onto the coating surface of the vehicle body through a programmable light source and use an industrial camera to capture 8 cosine phase images reflected from the coating surface of the vehicle body; Step 2: Use phase deflectometry to fuse the eight cosine phase images captured by the camera to generate a standard image, a specular reflection image, a diffuse reflection image, a wrapping phase image in the X direction, and a wrapping phase image in the Y direction; Step 3: Extract the gradient information of the wrapped phase image, fuse the obtained gradient information, and generate a shape map.

2. The method for detecting coating defects on complex vehicle body surfaces based on 2.5D phase deflectometry according to claim 1 is characterized in that: In step 1, the programmable light source projects four cosine phase images along the X direction and four cosine phase images along the Y direction, respectively, and the phase change between adjacent images is one quarter of a period.

3. The method for complex vehicle body surface coating defect detection based on 2.5D phase deflectometry according to claim 1 is characterized in that: In step 1, the eight cosine phase images captured by the industrial camera are all grayscale images.

4. The method for complex vehicle body surface coating defect detection based on 2.5D phase deflectometry according to claim 1 is characterized in that: In step 3, when generating the shape map, the gradient information of the wrapped phase map in the X and Y directions is extracted by applying the gradient operator, and then the obtained gradient values ​​are offset and modulo operations are performed to process the phase wrapping effect. Finally, the processed results are fused to generate the shape map.

5. The method for complex vehicle body surface coating defect detection based on 2.5D phase deflectometry according to claim 1 is characterized in that: In step 2, phase deflectometry is used to fuse the eight original cosine phase images to generate a standard image, a specular reflection image, a diffuse reflection image, a wrapped phase image in the X direction, and a wrapped phase image in the Y direction. The specific formula is as follows: Among them, I(x,y) represents the pixel gray value, A(x,y) represents the background light intensity, B(x,y) represents the modulated light intensity, represents phase and δ represents phase shift.

6. The method for complex vehicle body surface coating defect detection based on 2.5D phase deflectometry according to claim 5 is characterized in that: The standard image is generated by averaging and fusing eight original cosine phase images captured by an industrial camera.

7. The method for complex vehicle body surface coating defect detection based on 2.5D phase deflectometry according to claim 5 is characterized in that: The specular reflection map is generated by extracting the specular reflection components in the X and Y directions.

8. The method for complex vehicle body surface coating defect detection based on 2.5D phase deflectometry according to claim 5 is characterized in that: The diffuse map is generated by fusing the standard map and the specular map.

9. The method for complex vehicle body surface coating defect detection based on 2.5D phase deflectometry according to claim 5 is characterized in that: By solving the phase Generate a wrapped phase map in the X direction and a wrapped phase map in the Y direction.

10. The method for complex vehicle body surface coating defect detection based on 2.5D phase deflectometry according to claim 8, characterized in that: Using the diffuse reflection image, foreign matter and dirt on the surface can be detected using the following steps: S2.1, Image preprocessing: Use Gaussian kernel to perform convolution operation on the diffuse reflectance image to reduce random noise in the image and enhance the contrast of the image. Finally, the grayscale value of the image is normalized to 0 to 255 to obtain the processed image; S2.2, anomaly detection: calculating the gray value and standard deviation of the processed image and setting a dynamic detection threshold; using a sliding window to analyze local features and mark abnormal areas that exceed the threshold range; S2.3, feature extraction: perform connected domain analysis on the marked abnormal area, extract the morphological features of the abnormal area, and calculate the texture features of the abnormal area; S2.4, Classification analysis: Classify abnormal areas into foreign matter or dirt.

Images