Visual inspection method and system for processing quality of bicycle saddle

Abstract

The invention relates to the technical field of image processing and machine vision, in particular to a visual detection method and system for the machining quality of a bicycle saddle. Comprising the following steps: S1, acquiring a color image of a bicycle saddle, and separating three color channels of red, green and blue in the color image by taking a decorative small hole of the bicycle saddle as a center; s2, evaluating the continuous degree and the progressive degree in three color channels in the color image, and calculating the mixing degree of a local area through the continuous degree and the progressive degree based on the local area divided by taking each decorative small hole as the center; the continuity degree in each color channel refers to the smoothness and the similarity of pixel values in the same color channel. According to the invention, through high-resolution image processing and illumination normalization, stable and reliable detection is ensured; the continuity and the progressiveness of the color channel are evaluated, and the defect is accurately identified; and comparing regional color differences, and accurately positioning and evaluating defects in combination with time sequence analysis.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing and machine vision, and in particular to a visual inspection method and system for the processing quality of a bicycle saddle. Background Art

[0002] The quality of bicycle saddles directly impacts rider comfort and safety. High-quality saddles should exhibit excellent durability, shock absorption, and aesthetics. However, during the actual production process, surface defects such as cracks, scratches, and uneven color can appear on saddles. These issues not only affect the product's appearance and shorten its service life, but also pose potential safety risks. Therefore, accurate and efficient inspection of bicycle saddle processing quality is crucial.

[0003] Traditional manual visual inspection methods are inefficient and susceptible to subjective factors, making them difficult to meet the needs of large-scale production. With the development of computer vision technology, automated inspection methods based on image processing have gradually become mainstream. However, existing methods still have limitations when dealing with complex lighting conditions and varying color and texture features: images acquired under different lighting conditions can significantly affect inspection results. The diverse colors and textures of saddle surfaces increase the difficulty of defect identification. The presence of decorative small hole details makes it difficult to accurately assess quality using traditional image comparison methods. Summary of the Invention

[0004] In order to overcome the shortcomings of low efficiency and precision in saddle quality inspection, the present invention provides a method and system for visual inspection of bicycle saddle processing quality.

[0005] The technical solution of the present invention is: a visual inspection method for the processing quality of a bicycle saddle, comprising the following steps:

[0006] S1: Obtain a color image of a bicycle saddle, and separate the red, green, and blue color channels from the color image with the decorative hole of the bicycle saddle as the center.

[0007] S2: Evaluate the degree of continuity and progression within the three color channels of the color image. Based on the local area divided by each decorative hole as the center, calculate the degree of mixing of the local area using the degree of continuity and progression. The degree of continuity within the three color channels of the color image refers to the smoothness and similarity of pixel values ​​within the same color channel. The similarity refers to the difference in color values ​​between adjacent pixels. The degree of progression refers to the degree of consistency between the direction of change of color values ​​of adjacent pixels and the direction of change of the overall color value within the region within the same color channel.

[0008] S3: Divide the bicycle saddle into two areas, the left side is the detection area, and the right side is the verification area. The differences in the corresponding color channel values ​​in the two areas are compared to evaluate the degree of difference in the color channels in the corresponding areas.

[0009] S4: Determine a detection focus based on the mixing degree and the difference degree.

[0010] Preferably, the step of acquiring a color image of the bicycle saddle, taking the decorative small hole of the bicycle saddle as the center, and separating the red, green, and blue color channels in the color image comprises:

[0011] Get the light intensity value of the decorative hole area of ​​the bicycle saddle;

[0012] Taking each decorative hole as the center, extract the red, green, and blue color channel intensity values ​​of the hole area within a radius of 1 / 2 the distance from the center of the decorative hole to the nearest adjacent hole;

[0013] The intensity values ​​of the red, green, and blue color channels are divided by the light intensity value to obtain the first defect judgment ratio of the decorative hole area;

[0014] Under standard lighting conditions, a sample of defect-free saddles was collected, the first defect judgment ratio of each color channel was calculated, and the maximum value was taken as the preset threshold;

[0015] According to the preset threshold, the suspected defective area in the bicycle saddle area is analyzed during the light intensity change process.

[0016] Preferably, analyzing the suspected defective area of ​​the bicycle saddle area according to the preset threshold during the change of light intensity includes:

[0017] During the change of light intensity, the first defect judgment ratio of the entire bicycle saddle area is calculated;

[0018] If the three first defect judgment ratios of a certain area are all greater than the preset threshold, it is marked as a suspected defect area;

[0019] If the first defect judgment ratio of any channel in a certain area is less than a preset threshold, the area is not marked as a suspected defect area.

[0020] Preferably, the degree of continuity within each color channel refers to the smoothness and similarity of pixel values ​​within the same color channel, including:

[0021] Based on the suspected defect area, calculating the change values ​​of the red, green and blue channel values ​​in the suspected defect area and defining them as first change values;

[0022] At the same time, the change values ​​of the red, green and blue channel values ​​of the area adjacent to the suspected defect area are calculated and defined as the second change value;

[0023] If the first change value is not equal to the second change value, further calculating the change values ​​of the red, green, and blue channel values ​​of the adjacent area and defining them as the third change value;

[0024] The continuity degree of the suspected defect area is determined based on the first change value, the second change value, and the third change value.

[0025] Preferably, the degree of progression refers to the degree of consistency between the direction of change of the color values ​​of adjacent pixels and the direction of change of the overall color value in the region within the same color channel, including:

[0026] Based on the suspected defect areas, calculating the red, green, and blue channel values ​​of all suspected defect areas;

[0027] Based on the red, green, and blue channel values, calculating the degree of progression of the color channel values ​​between each suspected defect area and the nearest neighboring suspected defect area;

[0028] Starting from the rear end of the bicycle saddle, the progression of the suspected defect area from the rear end of the saddle to the front end of the saddle is obtained.

[0029] Preferably, the evaluating the degree of continuity and the degree of progression in the three color channels of the color image, based on the local area divided by each decorative hole as the center, calculates the degree of mixing of the local area by the degree of continuity and the degree of progression, including:

[0030] Obtain the union area of ​​the continuity degree and the progressive degree of the decorative small hole area; calculate the mixing degree through the mixing degree formula, which is as follows:

[0031]

[0032] in, The degree of mixing, For the The continuous area of ​​the decorative hole area; For the The progressive area of ​​the decorative aperture area; For the The degree of continuity of the decorative aperture area; For the The degree of progression of the decorative aperture area; is the number of decorative holes and the area of ​​the union; is the minimum value.

[0033] Preferably, the method of dividing the bicycle saddle into two areas, with the left side being a detection area and the right side being a verification area, and comparing the differences in corresponding color channel values ​​in the two areas to evaluate the degree of difference in color channels in the corresponding areas, includes:

[0034] Perform difference calculation on the three channel values ​​of the decorative small hole area in the detection area, specifically, calculate the difference between the maximum and minimum values ​​of the three channel values, and use the difference result as the difference degree of each area;

[0035] Based on the calculated difference degree of each area, the preset threshold is used to filter out the areas that exceed the preset threshold, and then compared with the verification area on the right;

[0036] Based on the comparison results, the overall degree of difference of the bicycle saddles was determined.

[0037] Preferably, determining the detection focus based on the mixing degree and the difference degree includes:

[0038] Get a bicycle saddle at Light intensity change curve at each moment;

[0039] Draw the decorative eyelet area of ​​the bicycle saddle Mixed degree curve and difference degree curve at each moment;

[0040] Draw a mean curve of the mixing degree curve and the difference degree curve;

[0041] The final defect area is determined by verifying the suspected defect area and adjacent areas through the verification formula.

[0042] Preferably, determining the final defect area of ​​the suspected defect area and the adjacent area by using a verification formula includes:

[0043] Based on the light intensity change curve, obtaining a light intensity change value;

[0044] Based on the mean curve, obtaining a mean change value;

[0045] The most defective area is determined by the verification formula, which is as follows:

[0046]

[0047] in, is the degree of defect, is the light intensity change value, is the mean change value.

[0048] A visual inspection system for the processing quality of a bicycle saddle, comprising:

[0049] Image acquisition and preprocessing module: This module uses a high-resolution camera to capture color images of bicycle saddles, separating the red, green, and blue color channels centered on the decorative holes. The intensity of each channel is normalized using the light intensity value to identify suspected defective areas.

[0050] Continuity and progression evaluation module: Based on the suspected defect area, calculate the changes in the red, green, and blue channel values ​​to determine the continuity and progression;

[0051] Difference degree assessment module: divides the saddle into a detection area and a verification area, compares the differences in color channel values ​​in the two areas, assesses the degree of difference, and screens out areas that exceed the preset threshold for comparative analysis;

[0052] Inspection focus determination module: Combine the degree of mixing and the degree of difference, draw the time curve of light intensity, degree of mixing and degree of difference, calculate the mean curve, and determine the final defect area through the verification formula.

[0053] Beneficial effects: The present invention, by introducing high-resolution color image processing and light intensity normalization technology, first solves the problem of inaccurate detection caused by changes in ambient light and color diversity in traditional detection methods, thereby ensuring the stability and reliability of the detection results. On this basis, by further evaluating the degree of continuity and progression within each color channel, an accurate analysis of the smoothness of the color transition on the saddle surface is achieved, thereby enhancing the detail of defect identification. Then, by comparing the differences in color channel values ​​in different areas of the saddle, potential processing defects are effectively screened out, thereby improving the comprehensiveness of the detection. Finally, combined with time series analysis and a comprehensive scoring mechanism, it is possible not only to accurately locate the defect position, but also to assess the severity, thereby providing a scientific basis for quality control in the production process. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 Flowchart of the visual inspection method for the processing quality of a bicycle saddle according to the present invention;

[0055] Figure 2 This is a system diagram for visual inspection of the processing quality of bicycle saddles according to the present invention. DETAILED DESCRIPTION

[0056] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0057] Example 1: A visual inspection method for the processing quality of a bicycle saddle, such as Figure 1As shown, the following steps are included:

[0058] S1: Obtain a color image of a bicycle saddle, and separate the red, green, and blue color channels from the color image with the decorative hole of the bicycle saddle as the center.

[0059] S2: Evaluate the degree of continuity and progression within the three color channels of the color image. Based on the local area divided by each decorative hole as the center, calculate the degree of mixing of the local area using the degree of continuity and progression. The degree of continuity within the three color channels of the color image refers to the smoothness and similarity of pixel values ​​within the same color channel. The similarity refers to the difference in color values ​​between adjacent pixels. The degree of progression refers to the degree of consistency between the direction of change of color values ​​of adjacent pixels and the direction of change of the overall color value within the region within the same color channel.

[0060] The adjacent pixels refer to directly adjacent pixels (such as up, down, left, right, or diagonal), emphasizing the close proximity of spatial positions; the adjacent pixels refer to pixels within a certain range (not limited to direct adjacency), focusing on the proximity within the area.

[0061] S3: Divide the bicycle saddle into two areas, the left side is the detection area, and the right side is the verification area. The differences in the corresponding color channel values ​​in the two areas are compared to evaluate the degree of difference in the color channels in the corresponding areas.

[0062] S4: Determine a detection focus based on the mixing degree and the difference degree.

[0063] Get a color image of a bicycle saddle and separate the red, green, and blue color channels from the image, centering on the decorative hole in the saddle. This includes:

[0064] Get the light intensity value of the decorative hole area of ​​the bicycle saddle;

[0065] Taking each decorative hole as the center, extract the red, green, and blue color channel intensity values ​​of the hole area within a radius of 1 / 2 the distance from the center of the decorative hole to the nearest adjacent hole;

[0066] The intensity values ​​of the red, green, and blue color channels are divided by the light intensity value to obtain the first defect judgment ratio of the decorative hole area;

[0067] Under standard lighting conditions, a sample of defect-free saddles was collected, the first defect judgment ratio of each color channel was calculated, and the maximum value was taken as the preset threshold;

[0068] According to the preset threshold, the suspected defective area in the bicycle saddle area is analyzed during the light intensity change process.

[0069] A further explanation is that a high-precision illuminance meter is used to measure the light intensity at the center of each decorative hole. Next, the intensity values ​​of the red, green, and blue color channels are extracted within a circular area with each decorative hole as the center and a radius equal to half the average distance between adjacent holes ( ). Subsequently, these intensity values ​​are divided by the light intensity value through normalization. , calculate the normalized , , .

[0070] According to the preset threshold, during the light intensity change process, the suspected defective area in the bicycle saddle area is analyzed, including:

[0071] During the change of light intensity, the first defect judgment ratio of the entire bicycle saddle area is calculated;

[0072] If the three first defect judgment ratios of a certain area are all greater than the preset threshold, the area is marked as a suspected defect area;

[0073] If the first defect judgment ratio of any channel in a certain area is less than a preset threshold, the area is not marked as a suspected defect area.

[0074] To further explain, if all first defect judgment ratios deviate from the preset threshold, it indicates that the color or reflective properties of the area have changed abnormally under the current lighting conditions, which is due to cracks, scratches or other surface defects in the bicycle saddle during processing. If all first defect judgment ratios still exceed the preset threshold under different lighting conditions, it indicates that there is an actual defect. If at least one first defect judgment ratio is within the preset threshold and does not deviate significantly from the expected color pattern, in this case, the color change in the area is considered normal and is caused by the texture of the material itself or slight color differences, rather than a processing defect.

[0075] The degree of continuity within each color channel refers to the smoothness and similarity of pixel values ​​within the same color channel, including:

[0076] Based on the suspected defect area, calculating the change values ​​of the red, green and blue channel values ​​in the suspected defect area and defining them as first change values;

[0077] At the same time, the change values ​​of the red, green and blue channel values ​​of the area adjacent to the suspected defect area are calculated and defined as the second change value;

[0078] If the first change value is not equal to the second change value, further calculating the change values ​​of the red, green, and blue channel values ​​of the adjacent area and defining them as the third change value;

[0079] The continuity degree of the suspected defect area is determined based on the first change value, the second change value, and the third change value.

[0080] A further explanation is that if there is a difference between the first change value and the second change value, this indicates that the saddle is defective and the degree of the defect is in a progressive relationship. In order to further confirm the severity of the defect, a third change value is introduced to verify the initial judgment. By calculating the changes in the color channel values ​​of adjacent areas, additional data support can be provided to help more accurately assess the existence and severity of defects, thereby ensuring the reliability of the detection results. The specific calculation method is that for each suspected defective area, the average values ​​of the red, green, and blue color channels are extracted, and the color differences (RGB Euclidean distances) of all adjacent pixel pairs in the suspected defective area are calculated, and the average value is taken to reflect the degree of color mutation within the area, which is defined as the first change value. ,

[0081]

[0082] The second change value is defined as the number of adjacent pixel pairs in the area, and the difference in the color channel mean between the suspected defect area and the directly adjacent area is calculated to determine whether the defect spreads outward. ,

[0083]

[0084] If the first change value and the second change value There are differences (i.e. ),in is the set threshold, and further calculates the difference between the adjacent area of ​​the adjacent area (the second layer of adjacent area) and the current adjacent area to verify the progressiveness of the defect.

[0085]

[0086] The third change value The color channel value difference between the adjacent area (i.e., the second layer adjacent area) and the current adjacent area is used to further verify the progressiveness of the defect. is the color channel mean of the current adjacent area, , , is the color channel mean of the adjacent area in the second layer.

[0087] The degree of progression refers to the degree of consistency between the direction of change of the color values ​​of adjacent pixels and the direction of change of the overall color value in the area within the same color channel, including:

[0088] Based on the suspected defect areas, calculating the red, green, and blue channel values ​​of all suspected defect areas;

[0089] Based on the red, green, and blue channel values, calculating the degree of progression of the color channel values ​​between each suspected defect area and the nearest neighboring suspected defect area;

[0090] Starting from the rear end of the bicycle saddle, the progression of the suspected defect area from the rear end of the saddle to the front end of the saddle is obtained.

[0091] A further explanation is that when there is a degree of progression between the suspected defect areas that are closest to each other, this means that the color change in the middle area is blocked or covered up, resulting in a misjudgment. In order to ensure the accuracy of the detection results, it is necessary to judge the degree of progression of these areas. If it is confirmed that the misjudgment is not caused by color change, but non-progressive changes caused by other factors, then these defects exist independently and need further analysis and confirmation. By judging the degree of progression between the suspected defect areas that are closest to each other, it is possible to effectively distinguish between true continuous defects and misjudgments caused by color changes. The specific calculation method is, for each suspected defect area, extract the average value of its red, green, and blue color channels; then, select the area with the closest pairwise distance, and use the Euclidean distance formula,

[0092]

[0093] The color difference between them is calculated, and the comprehensive difference between two adjacent suspected defect areas in the RGB color space is quantified by the Euclidean distance. Then, starting from the tail end of the bicycle saddle, the progressive degree of adjacent areas is calculated pair by pair and accumulated to obtain the total progressive degree.

[0094]

[0095] The color difference values ​​of adjacent areas are accumulated pair by pair from the rear end to the front end of the saddle to form a global progressive index.

[0096] Evaluate the degree of continuity and progression within the three color channels of the color image. Based on the local area divided by each decorative hole as the center, calculate the degree of mixing of the local area based on the degree of continuity and progression, including:

[0097] Obtain the union area of ​​the continuity degree and the progressive degree of the decorative small hole area; calculate the mixing degree through the mixing degree formula, which is as follows:

[0098]

[0099] in, The degree of mixing, For the The continuous area of ​​the decorative hole area; For the The progressive area of ​​the decorative aperture area; For the The degree of continuity of the decorative aperture area; For the The degree of progression of the decorative aperture area; is the number of decorative holes and the area of ​​the union; is the minimum value.

[0100] To further explain, assume that the total area of ​​a decorative small hole area is 10 square units, of which the continuous area accounts for 6 square units (60%) and the progressive area accounts for 4 square units (40%). In this case, the numerator is relatively large (that is, in the mixed degree formula, the value of the numerator is significantly higher than the denominator. It is specifically manifested in the following two scenarios: High coverage of continuity and progressive areas: When there are large areas of continuity-related areas (such as color mutation areas caused by cracks) and progressive-related areas (such as areas where color differences diffuse outward) in the decorative small hole area, and the union area of ​​the two almost covers the entire small hole, the numerator value will increase significantly. For example, if the continuity area of ​​a small hole is =6 cm 2 , progressive degree area =4 cm 2 , and the overlapping area between the two is 2cm 2 , then the numerator is 6+4-2=8 cm 2 The denominator is relatively small: if the continuous degree score and progressive level scoring Lower (such as and Close to the lower limit of the judgment threshold), then the ∑ and ∑ will remain small. For example, when =0.2, =0.3, the denominator is 0.2+e0.3≈0.2+1.35=1.55, at this time if the numerator is 8 cm 2 ,but ≈5.16, indicating that the molecular part is significantly dominant. Practical significance: The relatively large molecular part indicates that there are large-scale abnormal color changes (such as cracks or scratches) in the detection area, and these abnormal areas show both local mutations (high continuity) and diffusion to the surrounding areas (high progressiveness). At this time, the degree of mixing The value increases, indicating that the area needs to be prioritized for manual re-inspection or process adjustment to avoid potential quality defects. ), while the denominator (continuous degree value and progressive degree value after index adjustment) are 2 and 3 respectively. This leads to The value is large, about 1.96. A large M value indicates that there are significant color changes or potential defects in the area, which are caused by cracks and scratches during the processing, suggesting that further detailed inspection is needed to confirm the quality problem. Looking at another decorative hole area, the total area is also 10 square units, but the continuous degree area only accounts for 2 square units (20%), and the progressive degree area accounts for 8 square units (80%). In this case, the numerator is relatively small, while the denominator (the continuous degree value and the progressive degree value after index adjustment) are 8 and 7 respectively. This leads to The value is small, about 0.66. The value indicates that the color change is relatively smooth and there are no obvious defects, which indicates that the quality of the area is good and no further inspection is required. When calculating the degree of continuity, the first change value defined previously is used. and the second change value and the third change value , set a continuity judgment standard. For example, the continuity related area judgment: when the continuity index and exceeds the preset threshold (such as >α、 >β, and the difference between the two exceeds the allowable range (i.e. ), the third change value needs to be further calculated To verify the progressiveness of the defect. Also exceeding the verification threshold (e.g. >γ), the area is marked as a continuity-related area. The pixel area of ​​such an area in the decorative hole is counted by image analysis software and converted into the actual physical area based on the pre-calibrated resolution parameter (such as 0.1mm² per pixel) as the continuity area. Determination of areas related to the degree of progression: along the texture direction of the saddle surface (such as from the front to the rear end), calculate the color difference between adjacent suspected defect areas in sequence (using the Euclidean distance formula). If the color difference between multiple consecutive areas shows a monotonically increasing or decreasing trend (such as the absolute value of the linear regression slope k>δ), then mark the area as an area related to the degree of progression. Use image analysis tools to count the pixel area of ​​such areas, convert it into the actual area at the same resolution, and use it as the area of ​​the degree of progression. .

[0101] The bicycle saddle is divided into two areas: the left side is the detection area, and the right side is the verification area. The differences in the corresponding color channel values ​​in the two areas are compared to evaluate the degree of difference in the color channels in the corresponding areas, including:

[0102] Perform difference calculation on the three channel values ​​of the decorative small hole area in the detection area, specifically, calculate the difference between the maximum and minimum values ​​of the three channel values, and use the difference result as the difference degree of each area;

[0103] Based on the calculated difference degree of each area, the preset threshold is used to filter out the areas that exceed the preset threshold, and then compared with the verification area on the right;

[0104] Based on the comparison results, the overall degree of difference of the bicycle saddles was determined.

[0105] Further explanation: If the color changes in the inspection and verification areas at the same location are similar, these changes indicate actual defects. If they are inconsistent, it indicates a misjudgment due to ambient lighting, material texture, or other factors. This comparison further confirms or eliminates potential issues, ensuring the reliability of the test results. If the color changes in the inspection and verification areas are consistent and exceed a preset threshold, the saddle is considered to have an overall difference and a quality issue. If the color changes in most areas are within the preset threshold, the saddle is considered to be of good quality and no further inspection is required. This comparison comprehensively assesses the overall degree of difference in the saddle, providing more reliable test results.

[0106] Determining a detection focus based on the mixing degree and the difference degree includes:

[0107] Get a bicycle saddle at Light intensity change curve at each moment;

[0108] Draw the decorative eyelet area of ​​the bicycle saddle Mixed degree curve and difference degree curve at each moment;

[0109] Draw a mean curve of the mixing degree curve and the difference degree curve;

[0110] The final defect area is determined by verifying the suspected defect area and adjacent areas through the verification formula.

[0111] The final defect area is determined by verifying the suspected defect area and adjacent areas through the verification formula, including:

[0112] Based on the light intensity change curve, obtaining a light intensity change value;

[0113] Based on the mean curve, obtaining a mean change value;

[0114] The most defective area is determined by the verification formula, which is as follows:

[0115]

[0116] in, is the defect degree, is the value of light intensity change, is the mean change value.

[0117] For further explanation, a threshold T is set to distinguish the defect area and the normal area. If >T, then mark this area as a potential defect area. This indicates that the light change is relatively large while the quality change is small, which is a misjudgment caused by the change of ambient light. If <T, then it is considered that there is an actual defect in this area. This indicates that the quality change is significant while the light change is small, suggesting the existence of a real quality problem. If is approximately equal to T, then re-evaluate. The value of light intensity change is measured by using a high-precision illuminometer for the decorative small hole area of the bicycle saddle at different times , and then calculate the difference in light intensity between adjacent times, that is . For the mean change value , first calculate the mixing degree of the decorative small hole area at each time , and then calculate the difference in mixing degree between adjacent times, that is . When determining the threshold T, a large number of sample data are used for testing. Collect samples of bicycle saddles with different types and different quality conditions, and use this detection method for calculation and analysis. According to the distribution of the defect area and the normal area in the actual detection results, use statistical methods such as mean and standard deviation to determine a reasonable threshold T. For example, set it as the mean of the normal area plus a certain multiple of the standard deviation to ensure that the defect area and the normal area can be effectively distinguished in actual applications. Example 2: Based on Example 1, as

[0118] shown, a visual inspection system for the processing quality of a bicycle saddle includes: Figure 2 Image acquisition and preprocessing module: Use a high-resolution camera to obtain the color image of the bicycle saddle, separate the red, green, and blue color channels with the decorative small hole as the center; normalize the intensity of each channel through the light intensity value, and identify the suspected defect area;

[0119] Continuous degree and progressive degree evaluation module: Based on the suspected defect area, calculate the changes in the red, green, and blue channel values to determine the continuous degree and progressive degree;

[0120]

[0121] ​Difference degree assessment module: divides the saddle into a detection area and a verification area, compares the differences in color channel values ​​in the two areas, assesses the degree of difference, and screens out areas that exceed the preset threshold for comparative analysis;

[0122] Inspection focus determination module: Combine the degree of mixing and the degree of difference, draw the time curve of light intensity, degree of mixing and degree of difference, calculate the mean curve, and determine the final defect area through the verification formula.

[0123] The above is a detailed introduction to the present application. Specific examples are used herein to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method and core idea of ​​the present application. At the same time, for those skilled in the art, based on the idea of ​​the present application, there may be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A visual inspection method for the processing quality of bicycle saddles, characterized by: The following steps are involved: S1: Obtain a color image of a bicycle saddle, and separate the red, green, and blue color channels from the color image with the decorative hole of the bicycle saddle as the center. S2: Evaluate the degree of continuity and progression within the three color channels of the color image. Based on the local area divided by each decorative hole as the center, calculate the degree of mixing of the local area using the degree of continuity and progression. The degree of continuity within the three color channels of the color image refers to the smoothness and similarity of pixel values ​​within the same color channel. The similarity refers to the difference in color values ​​between adjacent pixels. The degree of progression refers to the degree of consistency between the direction of change of color values ​​of adjacent pixels and the direction of change of the overall color value within the region within the same color channel. S3: Divide the bicycle saddle into two areas, the left side is the detection area, and the right side is the verification area. The differences in the corresponding color channel values ​​in the two areas are compared to evaluate the degree of difference in the color channels in the corresponding areas. S4: Determine a detection focus based on the mixing degree and the difference degree.

2. The visual inspection method for the processing quality of a bicycle saddle according to claim 1, characterized in that: The method of obtaining a color image of a bicycle saddle, taking the decorative small hole of the bicycle saddle as the center, and separating the red, green, and blue color channels in the color image includes: Get the light intensity value of the decorative hole area of ​​the bicycle saddle; Taking each decorative hole as the center, extract the red, green, and blue color channel intensity values ​​of the hole area within a radius of 1 / 2 the distance from the center of the decorative hole to the nearest adjacent hole; The intensity values ​​of the red, green, and blue color channels are divided by the light intensity value to obtain the first defect judgment ratio of the decorative hole area; Under standard lighting conditions, defect-free saddle samples are collected, the first defect judgment ratio of each color channel is calculated, and the maximum value is taken as the preset threshold; based on the preset threshold, the suspected defect area in the bicycle saddle area is analyzed during the change of light intensity.

3. The visual inspection method for the processing quality of a bicycle saddle according to claim 2, characterized in that: The step of analyzing the suspected defective area of ​​the bicycle saddle according to the preset threshold during the light intensity change process includes: During the change of light intensity, the first defect judgment ratio of the entire bicycle saddle area is calculated; If the three first defect judgment ratios of a certain area are all greater than the preset threshold, it is marked as a suspected defect area; If the first defect judgment ratio of any channel in a certain area is less than a preset threshold, the area is not marked as a suspected defect area.

4. The visual inspection method for the processing quality of a bicycle saddle according to claim 1, wherein: The degree of continuity within each color channel refers to the smoothness and similarity of pixel values ​​within the same color channel, including: Based on the suspected defect area, calculating the change values ​​of the red, green and blue channel values ​​in the suspected defect area and defining them as first change values; At the same time, the change values ​​of the red, green and blue channel values ​​of the area adjacent to the suspected defect area are calculated and defined as the second change value; If the first change value is not equal to the second change value, further calculating the change values ​​of the red, green, and blue channel values ​​of the adjacent area and defining them as the third change value; The continuity degree of the suspected defect area is determined based on the first change value, the second change value, and the third change value.

5. The visual inspection method for the processing quality of a bicycle saddle according to claim 1, characterized in that: The degree of progression refers to the degree of consistency between the direction of change of the color values ​​of adjacent pixels and the direction of change of the overall color value in the region within the same color channel, including: Based on the suspected defect areas, calculating the red, green, and blue channel values ​​of all suspected defect areas; Based on the red, green, and blue channel values, calculating the degree of progression of the color channel values ​​between each suspected defect area and the nearest neighboring suspected defect area; Starting from the rear end of the bicycle saddle, the progression of the suspected defect area from the rear end of the saddle to the front end of the saddle is obtained.

6. The visual inspection method for the processing quality of a bicycle saddle according to claim 1, characterized in that: The evaluation of the degree of continuity and the degree of progression in the three color channels of the color image is based on the local area divided by each decorative hole as the center, and the calculation of the degree of mixing of the local area by the degree of continuity and the degree of progression includes: Obtain the union area of ​​the continuity degree and the progressive degree of the decorative small hole area; calculate the mixing degree through the mixing degree formula, which is as follows: ； in, The degree of mixing, For the The continuous area of ​​the decorative hole area; For the The progressive area of ​​the decorative aperture area; For the The degree of continuity of the decorative aperture area; For the The degree of progression of the decorative aperture area; is the number of decorative holes and the area of ​​the union; is the minimum value.

7. The visual inspection method for the processing quality of a bicycle saddle according to claim 1, characterized in that: The bicycle saddle is divided into two areas, the left side is the detection area, and the right side is the verification area, and the differences in the corresponding color channel values ​​in the two areas are compared to evaluate the degree of difference in the color channels in the corresponding areas, including: Perform difference calculation on the three channel values ​​of the decorative small hole area in the detection area, specifically, calculate the difference between the maximum and minimum values ​​of the three channel values, and use the difference result as the difference degree of each area; Based on the calculated difference degree of each area, the preset threshold is used to filter out the areas that exceed the preset threshold, and then compared with the verification area on the right; Based on the comparison results, the overall degree of difference of the bicycle saddles was determined.

8. The visual inspection method for the processing quality of a bicycle saddle according to claim 1, characterized in that: The determining of the detection focus based on the mixing degree and the difference degree includes: Get a bicycle saddle at Light intensity change curve at each moment; Draw the decorative eyelet area of ​​the bicycle saddle Mixed degree curve and difference degree curve at each moment; Draw a mean curve of the mixing degree curve and the difference degree curve; The final defect area is determined by verifying the suspected defect area and adjacent areas through the verification formula.

9. The visual inspection method for the processing quality of a bicycle saddle according to claim 8, characterized in that: The final determination of the defect area by using a verification formula for the suspected defect area and the adjacent area includes: Based on the light intensity change curve, obtaining a light intensity change value; Based on the mean curve, obtaining a mean change value; The most defective area is determined by the verification formula, which is as follows: ； in, is the degree of defect, is the light intensity change value, is the mean change value.

10. A visual inspection system for the processing quality of bicycle saddles, used to implement the visual inspection method for the processing quality of bicycle saddles according to any one of claims 1 to 9, characterized in that: include: Image acquisition and preprocessing module: This module uses a high-resolution camera to capture color images of bicycle saddles, separating the red, green, and blue color channels centered on the decorative holes. The intensity of each channel is normalized using the light intensity value to identify suspected defective areas. Continuity and progression evaluation module: Based on the suspected defect area, calculate the changes in the red, green, and blue channel values ​​to determine the continuity and progression; Difference degree assessment module: divides the saddle into a detection area and a verification area, compares the differences in color channel values ​​in the two areas, assesses the degree of difference, and screens out areas that exceed the preset threshold for comparative analysis; Inspection focus determination module: Combine the degree of mixing and the degree of difference, draw the time curve of light intensity, degree of mixing and degree of difference, calculate the mean curve, and determine the final defect area through the verification formula.

Images