A method for detecting the eligibility of a thin-walled circular sleeve based on image processing technology

Abstract

The application discloses a kind of thin-walled round sleeve eligibility detection methods based on image processing technology, specific steps are as follows: first, image is binarized and contour detection, then the two profiles of most and second most profile points are screened and circle fitting is carried out, and then the inner and outer profiles of thin-walled round sleeve are determined according to the fitting circle radius;Select the center of the fitting circle with the smallest difference from the standard circle in the fitting circle of the inner and outer profiles, create a ray with the center as the starting point and calculate the wall thickness difference of the thin-walled round sleeve;Finally, according to the inner profile angle of the thin-walled round sleeve and the distance between the profile point and the center of the fitting circle, it is judged whether there is burr on the inside of the thin-walled round sleeve, and whether the measured thin-walled round sleeve is qualified is judged according to whether the maximum wall thickness difference of the thin-walled round sleeve is out of tolerance and whether there is burr. The method can quickly and accurately determine whether the thin-walled round sleeve part is qualified, and improves the detection efficiency.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of image processing technology, and particularly relates to a thin-walled round sleeve qualification detection method based on image processing technology. BACKGROUND

[0002] In the production process of the factory pipeline, the quality detection link is particularly important. The quality detection in multiple process links can effectively guarantee the qualified rate of the production parts, and also reduces the production cost.

[0003] The traditional quality detection of the thin-walled round sleeve workpiece is mostly through the workers using the dial gauge and the vernier caliper to measure the size and observing the burrs by the naked eyes, so the detection efficiency is low, and the detection quality is greatly affected by the human factors. For the industrial production activities pursuing high efficiency, the traditional method cannot be well applied to the pipeline, and more intelligent devices should be combined to reduce the human intervention to improve the product quality detection efficiency and effect. But so far, the research results of the thin-walled round sleeve qualification automatic detection method at home and abroad are very few. SUMMARY

[0004] In view of the deficiencies in the background art, the purpose of the present application is to provide a thin-walled round sleeve qualification detection method based on image processing technology, to improve the measurement accuracy and detection efficiency.

[0005] To achieve the above purpose, the technical scheme adopted by the present application is a thin-walled round sleeve qualification detection method based on image processing technology, which is implemented according to the following steps:

[0006] Step 1: performing adaptive threshold binarization processing on the original thin-walled round sleeve gray image;

[0007] Step 2: performing contour detection in the binary image to obtain two contours with the most and the second most contour points;

[0008] Step 3: performing circle fitting on the two contours obtained in step 2 to determine the inner contour and the outer contour of the thin-walled round sleeve;

[0009] Step 4: selecting the contour with a small difference from the standard circle from the inner contour and the outer contour of the thin-walled round sleeve, and obtaining the center O of the fitting circle corresponding to the contour;

[0010] Step 5: taking the center O obtained in step 4 as the starting point to make a ray outward, calculating the wall thickness of the thin-walled round sleeve in the direction of the ray, and rotating the ray by a fixed value step angle and calculating the wall thickness of the thin-walled round sleeve in different directions of the rays respectively, to obtain the maximum wall thickness difference of the thin-walled round sleeve;

[0011] Step 6: Calculate the angle between the line connecting each profile point of the inner profile of the thin-walled circular sleeve and the center of the fitted circle of the inner profile and the horizontal line, and the distance from each profile point of the inner profile of the thin-walled circular sleeve to the center of the fitted circle of the inner profile, and determine whether burrs exist on the inner side of the thin-walled circular sleeve according to the obtained angle and distance.

[0012] Step 7: Determine whether the thin-walled circular sleeve is qualified according to whether the maximum wall thickness difference of the thin-walled circular sleeve is out of tolerance and whether burrs exist on the inner side of the thin-walled circular sleeve.

[0013] Further, the two profiles with the most and the second most profile points are obtained by the following steps: 1) performing profile detection on the image after binaryzation; 2) counting the number of profile points of each profile; 3) sorting all profiles according to the number of profile points; and 4) extracting the two profiles with the most and the second most profile points.

[0014] Further, the method for determining the inner profile and the outer profile of the thin-walled circular sleeve is as follows: performing circle fitting on the two profiles obtained in step 2 by using the least square method, and the profile corresponding to the fitted circle with the larger diameter is the outer profile of the thin-walled circular sleeve, and the profile corresponding to the fitted circle with the smaller diameter is the inner profile of the thin-walled circular sleeve.

[0015] Further, the method for selecting the profile with small difference from a circle in step 4 is as follows: calculating the circular degree e of the inner profile and the outer profile of the thin-walled circular sleeve respectively, and the profile with the circular degree closest to 1 is the profile with small difference from a circle.

[0016] The formula for calculating the circular degree e is as follows:

[0017] e = 4πS / L 2

[0018] In the formula, S is the area of the region enclosed by the profile, i.e. the number of pixels enclosed by the profile, and L is the perimeter of the profile, i.e. the number of profile points of the profile. When the circular degree e is 1, the profile is a standard circle, and the smaller the e is, the more irregular the profile is, and the greater the difference from the standard circle is.

[0019] Further, the method for obtaining the wall thickness of the thin-walled circular sleeve in a certain direction of a ray is as follows:

[0020] Taking the center O of the fitted circle obtained in step 4 as the starting point and another arbitrary point Q0, a ray is drawn from the point Q0 to intersect the inner profile fitting circle and the outer profile fitting circle of the thin-walled circular sleeve at points P1 and P2, the point P3 closest to P1 is found on the inner profile, and the point P4 closest to P2 is found on the outer profile, and the distance between P3 and P4 is the wall thickness d0 of the thin-walled circular sleeve in the direction of the ray.

[0021] Further, the method for obtaining the maximum wall thickness difference of the thin-walled circular sleeve is as follows:

[0022] The point Q0(x0, y0) is rotated around the center O(x, y) of the fitted circle by an angle of 90°, and the point Q1(x1, y1) obtained after rotation is taken as the starting point of the ray, and a ray is drawn from the point Q1 to intersect the inner profile fitting circle and the outer profile fitting circle of the thin-walled circular sleeve at points P5 and P6, the point P7 closest to P5 is found on the inner profile, and the point P8 closest to P6 is found on the outer profile, and the distance between P7 and P8 is the wall thickness d1 of the thin-walled circular sleeve in the direction of the ray. c ,yc Rotate by an integer multiple of a fixed step angle θ to obtain the rotated point Q0. i (x i ,y i Given i = 1, 2, ..., n-1, calculate ray OQ. i thin-walled circular sleeve wall thickness d in the direction i Find the maximum thin-walled circular sleeve wall thickness d among n thin-walled circular sleeve wall thicknesses. max and minimum thin-walled circular sleeve wall thickness d min Then the maximum wall thickness difference of the thin-walled circular sleeve is d = d max -d min .

[0023] The formula for calculating the step angle θ is: θ = 360° / n;

[0024] The point Q i The formula set for solving the coordinates is as follows:

[0025] x i =(x0-x c cos(iθ)-(y0-y c sin(iθ)+x c

[0026] y i =(x0-x c sin(iθ)-(y0-y c cos(iθ)+y c

[0027] In the formula, i = 1, 2, ..., n-1.

[0028] Furthermore, determining whether there are burrs on the inner side of the thin-walled cylindrical sleeve mainly includes the following steps:

[0029] Fit the center O of the circle from the inner contour of the thin-walled circular sleeve. in (x in ,y in Starting from the right horizontal direction, continuously obtain the contour points A of the inner contour of the thin-walled circular sleeve in a clockwise or counterclockwise direction. j (x j ,y j ), j = 1, 2, ..., m, m is the number of contour points of the inner contour of the thin-walled circular sleeve;

[0030] Calculate the contour points A of the inner contour of the thin-walled circular sleeve. j To the center O in Distance D j :

[0031] D j =[(x j -xin ) 2 +(y j -y in ) 2 ] 1 / 2

[0032] In the formula, j = 1, 2, ..., m.

[0033] Calculate the distance from all contour points of the inner contour of the thin-walled circular sleeve to the center O. in The average distance D avg :

[0034] D avg = (D1+D2+…+D m ) / m

[0035] Calculate the center O in Contour points A of the inner contour of the thin-walled circular sleeve j The angle α between the line connecting the two sides and the horizontal direction j α j The calculation formula is as follows:

[0036] α j =arctan[(y j -y in ) / (x j -x in )]

[0037] In the formula, j = 1, 2, ..., m.

[0038] Furthermore, the specific conditions for determining whether there are burrs on the inner side of the thin-walled circular sleeve are as follows:

[0039] Condition 1: Angle α j Continuously increasing or continuously decreasing, j = 1, 2, ..., m;

[0040] Condition 2: Distances D j All are within the interval (D) avg -th1,D avg Within (+th2), th1 and th2 are two real numbers that can be determined experimentally, j = 1, 2, ..., m;

[0041] If both conditions 1 and 2 are met, it is determined that there are no burrs on the inside of the thin-walled circular sleeve; otherwise, it is determined that there are burrs on the inside of the thin-walled circular sleeve.

[0042] Furthermore, the criteria for determining whether a thin-walled round sleeve is qualified are as follows:

[0043] Condition 3: The maximum wall thickness difference d of the thin-walled circular sleeve is less than the threshold th3, which is determined experimentally;

[0044] Condition 4: There are no burrs on the inner side of the thin-walled round sleeve;

[0045] When condition 3 and condition 4 are satisfied at the same time, then the thin-wall round sleeve is determined to be qualified; otherwise, the thin-wall round sleeve is determined to be unqualified.

[0046] Compared with the background art, the gain effect of the present application is:

[0047] 1. The present application realizes non-contact automatic measurement of the wall thickness of the thin-wall round sleeve by using image processing technology and can detect the inner burr, so as to effectively avoid false detection caused by errors of instruments and human factors compared with artificial detection.

[0048] 2. The present application measures the wall thickness after circular fitting of the extracted contour, uses the circular degree of the fitted circle to compare the fitting effect to screen out the best fitting circle center, and uses the actual contour point closest to the fitting point when measuring the wall thickness, so as to make the wall thickness measurement of the thin-wall round sleeve more accurate.

[0049] 3. The judgment of the inner burr of the thin-wall round sleeve is realized by comparing the relationship between the horizontal direction included angles of the contour points and the line connecting the inner circle center, so compared with the method of only detecting the distance between the contour point and the circle center, the present method can more accurately judge whether the inner side of the thin-wall round sleeve exists burr. BRIEF DESCRIPTION OF DRAWINGS

[0050] Figure 1 is the overall flow chart of the thin-wall round sleeve wall thickness measurement of the present application;

[0051] Figure 2 is the gray scale image of the end face of the thin-wall round sleeve to be detected of the present application;

[0052] Figure 3 is the binary image of the end face of the thin-wall round sleeve after preprocessing of the present application;

[0053] Figure 4 are two contour images of the end face of the thin-wall round sleeve of the present application;

[0054] Figure 5 is the inner and outer contour fitting circle image of the present application;

[0055] Figure 6 is the wall thickness calculation schematic diagram of the thin-wall round sleeve of the present application;

[0056] Figure 7 is the ray generation schematic diagram of the present application;

[0057] Figure 8 is the actual position image of the maximum and minimum wall thickness of the present application;

[0058] Figure 9 is the included angle calculation method schematic diagram of the present application;

[0059] Figure 10The burr actual position image of the present application. DETAILED DESCRIPTION

[0060] The present application is further described below in conjunction with the accompanying drawings and specific embodiments (examples), but is not limited to the basis of the application.

[0061] It should be noted that the present application uses an industrial camera and a lens to collect the end surface gray scale image of the thin-walled round sleeve part by backlighting with a surface light source.

[0062] As shown in Figure 1 , the steps of the method of the present application are:

[0063] Step 1: Adaptive threshold binarization processing of the original thin-walled round sleeve gray scale image

[0064] Read the collected thin-walled round sleeve original image as shown in Figure 2 , and use the OTSU method to automatically obtain the threshold value to perform binarization processing on the thin-walled round sleeve end surface image. The processed image has a black background and a white target, as shown in Figure 3 .

[0065] Step 2: Contour detection in the binary image to obtain two contours with the most and the second most contour points

[0066] Contour detection is performed on the binary image, and the number of contour points included in each contour is calculated. The contours are sorted according to the number of contour points, and the two contours with the most and the second most contour points, Con1 and Con2, are selected, as shown in Figure 4 . These two contours are the inner and outer contours of the thin-walled round sleeve.

[0067] Step 3: Circular fitting of the two contours obtained in Step 2 to determine the inner and outer contours of the thin-walled round sleeve

[0068] Least squares method is used to perform circular fitting on the two contours Con1 and Con2, and two fitted circles C1 and C2 are obtained, as shown in Figure 5 . Since there may be burrs on the inner and outer walls of the thin-walled round sleeve, the correspondence between the inner and outer circles and the contours cannot be determined by the sorting result of the number of contour points. Therefore, the contour corresponding to the larger diameter fitted circle is determined as the outer contour of the thin-walled round sleeve, and the contour corresponding to the smaller diameter fitted circle is determined as the inner contour of the thin-walled round sleeve.

[0069] Step 4: Selecting the contour with a small difference from the circle among the inner and outer contours of the thin-walled round sleeve, and obtaining the center O of the fitted circle corresponding to the contour

[0070] Since the rays are needed to calculate the wall thickness of the thin-walled round sleeve part in the following step 5, the center O of the fitting circle corresponding to the profile of the roundness e closer to the standard circle is selected as the starting point of the rays. The formula for calculating the roundness e is:

[0071] e = 4πS / L 2

[0072] In the formula, S is the area of the region enclosed by the profile, that is, the number of pixels enclosed by the profile, and L is the perimeter of the profile, that is, the number of profile points of the profile. When the roundness e is 1, the profile is a standard circle. The smaller the roundness e, the more irregular the profile shape, and the greater the difference from the roundness.

[0073] Step 5: A ray is drawn outward from the center O obtained in step 4, and the wall thickness of the thin-walled round sleeve in the direction of the ray is calculated. Meanwhile, the ray is rotated by a constant step angle, and the wall thickness of the thin-walled round sleeve in different directions of the rays is calculated respectively, so as to obtain the maximum wall thickness difference of the thin-walled round sleeve

[0074] After the starting point O is obtained, the rays used to calculate the pixel value of the wall thickness need to be drawn, as shown in Figure 6 The ray OQ0 intersects the inner and outer circles at P 01 , P 02 , and the nearest pixel point P 01 to the intersection point P 03 on the end surface inner contour of the thin-walled round sleeve part is found, and the nearest pixel point P 04 to the intersection point P 02 on the outer contour is found. The Euclidean distance between P 03 and P 04 is the pixel value d0 of the wall thickness of the thin-walled round sleeve part in the direction of the ray OQ0.

[0075] As shown in Figure 7 , the ray OQ0 is rotated counterclockwise by an angle θ with O as the rotation center to obtain the ray OQ1. The pixel value d1 of the wall thickness of the thin-walled round sleeve part in the direction of the ray OQ1 is calculated according to the above method.

[0076] Continue to rotate the rays and calculate the pixel value of the wall thickness of the thin-walled round sleeve part in the direction of the rotated rays according to the above steps until the ray OQ n-1 and the pixel value d n-1 of the wall thickness of the thin-walled round sleeve part are obtained, where n = 360° / θ.

[0077] The coordinate solving formula group of the point Q i is as follows:

[0078] x i = (x0-x c )cos(iθ)-(y0-y c )sin(iθ)+xc

[0079] y i =(x0-x c sin(iθ)-(y0-y c cos(iθ)+y c

[0080] In the formula, i = 1, 2, ..., n-1.

[0081] Find the maximum pixel value d of the wall thickness of the thin-walled circular sleeve part in all ray directions. max Minimum value d min and average value d avg The pixel value of the wall thickness of the thin-walled circular sleeve part is the average value d. avg The pixel value of the wall thickness difference is d max With d min The difference.

[0082] Then, the actual physical quantity represented by each pixel is obtained through calibration, that is, the pixel equivalent.

[0083] By multiplying the pixel values ​​of the maximum wall thickness, minimum wall thickness, and wall thickness of the thin-walled circular sleeve obtained in the above steps by the pixel equivalent, the actual values ​​of the maximum wall thickness, minimum wall thickness, and wall thickness of the thin-walled circular sleeve can be obtained.

[0084] like Figure 8 As shown, "1" is the thinnest part of the thin-walled circular sleeve, and "2" is the thickest part of the thin-walled circular sleeve.

[0085] Step 6: Calculate the angle between the line connecting each contour point of the inner contour of the thin-walled circular sleeve and the center of the fitted circle of the inner contour, and the horizontal line, as well as the distance from each contour point of the inner contour of the thin-walled circular sleeve to the center of the fitted circle of the inner contour. Based on the obtained angles and distances, determine whether there are burrs on the inner side of the thin-walled circular sleeve.

[0086] The calculation of the angle between the line connecting each contour point of the inner contour of the thin-walled circular sleeve and the center of the fitted circle of the inner contour and the horizontal line is shown in the figure below. Figure 9 As shown. The required included angle is at the center O of the circle. in Contour points A of the inner contour of the thin-walled circular sleeve j The angle α between the line connecting the two points and the negative half-axis of the horizontal y-axis. j The calculation formula is as follows:

[0087] α j =arctan[(y j -y in ) / (x j -x in )]+β

[0088] where β is a correction value, β = 180° when the profile point falls in the first and fourth quadrants, β = 360° when the profile point falls in the second quadrant, and β = 0° in other cases. When x j = x in , the profile point falls on the positive half of the y-axis, and α j = 180°; when the profile point falls on the negative half of the y-axis, α j = 0°.

[0089] The distance of each profile point of the inner profile of the thin-walled circular sleeve to the center of the fitting circle of the inner profile is calculated. The calculation formula of the distance D j of each profile point A in of the inner profile of the thin-walled circular sleeve to the center O j of the circle is as follows:

[0090] D j = [(x j -x in ) 2 +(y j -y in ) 2 ] 1 / 2

[0091] where j = 1, 2, …, m, and m is the number of profile points of the inner profile of the thin-walled circular sleeve.

[0092] The average value D avg of the distance of all profile points of the inner profile of the thin-walled circular sleeve to the center O in of the circle is calculated.

[0093] D avg = (D1+D2+…+D m ) / m

[0094] Whether there is burr on the inner side of the thin-walled circular sleeve is determined by the following conditions:

[0095] Condition 1: the angle α j is continuously increasing or continuously decreasing, j = 1, 2, …, m.

[0096] Condition 2: each distance D j is within the interval (D avg -th1, D avg +th2), th1 and th2 are two real numbers, which can be determined by experiments, j = 1, 2, …, m.

[0097] After D avg and a series of α j , D j are obtained, appropriate th1 and th2 are selected according to actual requirements. When conditions 1 and 2 are both satisfied, it is determined that there is no burr on the inner side of the thin-walled circular sleeve; otherwise, it is determined that there is burr on the inner side of the thin-walled circular sleeve.Figure 10 For the detection result of this embodiment, burrs exist at the gray marked positions of "3" and "4".

[0098] Step 7: According to whether the maximum wall thickness difference of the thin-walled round sleeve exceeds the threshold and whether burrs exist on the inner side of the thin-walled round sleeve, it is determined whether the thin-walled round sleeve is qualified

[0099] The condition for determining whether the thin-walled round sleeve is qualified is:

[0100] Condition 3: The maximum wall thickness difference d of the thin-walled round sleeve is less than a threshold th3, which is determined by experiment.

[0101] Condition 4: No burrs exist on the inner side of the thin-walled round sleeve.

[0102] When conditions 3 and 4 are both satisfied, the measured thin-walled round sleeve is a qualified piece, otherwise it is an unqualified piece.

[0103] The above is only a preferred specific embodiment of the present application, but the protection scope of the present application is not limited to this. Any person skilled in the art can easily think of changes or replacements within the technical range disclosed by the present application, which should be covered in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method for detecting the conformity of thin-walled circular sleeves based on image processing technology, characterized in that, The specific steps are as follows: Step 1: Perform adaptive thresholding binarization on the original thin-walled circular grayscale image; Step 2: Perform contour detection in the binary image and obtain the two contours with the most and second most contour points; Step 3: Perform circle fitting on the two contours obtained in Step 2 to determine the inner and outer contours of the thin-walled circular sleeve; Step 4: Select the contour with the smallest difference from the standard circle from the inner and outer contours of the thin-walled circular sleeve, and obtain the center O of the fitted circle corresponding to the contour. Step 5: Draw a ray outward from the center O obtained in Step 4, calculate the wall thickness of the thin-walled sleeve in the ray direction, and rotate the ray by a fixed step angle and calculate the wall thickness of the thin-walled sleeve in different ray directions respectively, so as to obtain the maximum wall thickness difference of the thin-walled sleeve. Step 6: Calculate the angle between the line connecting each contour point of the inner contour of the thin-walled circular sleeve and the center of the inner contour fitting circle and the horizontal line, as well as the distance from each contour point of the inner contour of the thin-walled circular sleeve to the center of the inner contour fitting circle. Based on the obtained angle and distance, determine whether there are burrs on the inner side of the thin-walled circular sleeve. Step 7: Determine whether the thin-walled sleeve is qualified based on whether the maximum wall thickness difference of the thin-walled sleeve exceeds the tolerance and whether there are burrs on the inner side of the thin-walled sleeve. Step 6, determining whether there are burrs on the inner side of the thin-walled sleeve, mainly includes the following steps: Step 61: Fit the center O of the circle from the inner contour of the thin-walled circular sleeve. in (x in ,y in Starting from the right horizontal direction, continuously obtain the contour points A of the inner contour of the thin-walled circular sleeve in a clockwise or counterclockwise direction. j (x j ,y j ), j = 1, 2, ..., m, m is the number of contour points of the inner contour of the thin-walled circular sleeve; Step 62: Calculate the contour points A of the inner contour of the thin-walled circular sleeve. j To the center O in Distance D j : D j =[(x j -x in ) 2 +(y j -y in ) 2 ] 1 / 2 In the formula, j = 1, 2, ..., m; Step 63: Calculate the distance from all contour points of the inner contour of the thin-walled circular sleeve to the center O. in The average distance D avg ; D avg =(D1+D2+…+D m ) / m Step 64: Calculate the center O of the circle in Contour points A of the inner contour of the thin-walled circular sleeve j The angle α between the line connecting the two sides and the horizontal direction j α j The calculation formula is as follows: α j =arctan[(and j -and in ) / (x j -x in )]+β In the formula, j = 1, 2, ..., m; β is the correction value; Step 65: Determine whether there are burrs on the inside of the thin-walled round sleeve using the following conditions: Condition 1: Angle α j Continuously increasing or continuously decreasing, j = 1, 2, ..., m; Condition 2: Distances D j All are within the interval (D) avg -th1,D avg Within (+th2), th1 and th2 are two real numbers, determined experimentally, j = 1, 2, ..., m; If both conditions 1 and 2 are met, it is determined that there are no burrs on the inside of the thin-walled circular sleeve; otherwise, it is determined that there are burrs on the inside of the thin-walled circular sleeve.

2. The method for detecting the conformity of thin-walled circular sleeves based on image processing technology according to claim 1, characterized in that: Step 2, obtaining the two contours with the most and second-most contour points, mainly includes the following steps: Step 21: Perform contour detection in the binarized image; Step 22: Count the number of contour points for each contour; Step 23: Sort all contours according to the number of contour points; Step 24: Extract the two contours with the most and second most points.

3. The method for detecting the conformity of thin-walled circular sleeves based on image processing technology according to claim 1, characterized in that: In step 3, the method for determining the inner and outer contours of the thin-walled circular sleeve is as follows: the two contours obtained in step 2 are fitted with circles using the least squares method. The contour corresponding to the fitted circle with the larger diameter is the outer contour of the thin-walled circular sleeve, and the contour corresponding to the fitted circle with the smaller diameter is the inner contour of the thin-walled circular sleeve.

4. The method for detecting the conformity of thin-walled circular sleeves based on image processing technology according to claim 1, characterized in that: In step 4, the method for selecting the contour with the smallest difference from the circle is as follows: calculate the roundness e of the inner and outer contours of the thin-walled circular sleeve respectively, and the contour with the roundness closest to 1 is the contour with the smallest difference from the circle. The formula for calculating the roundness e is: e = 4πS / L 2 In the formula, S is the area of ​​the region enclosed by the contour, that is, the number of pixels enclosed by the contour; L is the perimeter of the contour, that is, the number of contour points of the contour.

5. The method for detecting the conformity of thin-walled circular sleeves based on image processing technology according to claim 1, characterized in that: In step 5, the method for determining the wall thickness of the thin-walled circular sleeve along a certain ray direction is as follows: taking the center O obtained in step 4 as the starting point and taking any other point Q0, draw a ray towards point Q0, intersecting the inner contour fitting circle and the outer contour fitting circle of the thin-walled circular sleeve at points P1 and P2 respectively. Find the point P3 closest to P1 on the inner contour and the point P4 closest to P2 on the outer contour. The distance between P3 and P4 is the wall thickness d0 of the thin-walled circular sleeve along the ray direction.

6. The method for detecting the conformity of thin-walled circular sleeves based on image processing technology according to claim 1, characterized in that: In step 5, the method for determining the maximum wall thickness difference of the thin-walled circular sleeve is as follows: Point Q0(x0,y0) is rotated around the center O(x0,y0) of the circle. c ,y c Rotate by an integer multiple of a fixed step angle θ to obtain the rotated point Q0. i (x i ,y i Given i = 1, 2, ..., n-1, calculate ray OQ. i thin-walled circular sleeve wall thickness d in the direction i Find the maximum thin-walled circular sleeve wall thickness d among n thin-walled circular sleeve wall thicknesses. max and minimum thin-walled circular sleeve wall thickness d min Then the maximum wall thickness difference of the thin-walled circular sleeve is d = d max -d min ; The formula for calculating the step angle θ is: θ = 360° / n; The point Q i The formula set for solving the coordinates is as follows: x i =(x0-x c )cos(iθ)-(y0-y c )sin(iθ)+x c and i =(x0-x c )sin(iθ)-(y0-y c )cos(iθ)+y c In the formula, i = 1, 2, ..., n-1.

7. The method for detecting the conformity of thin-walled circular sleeves based on image processing technology according to claim 1, characterized in that: In step 6, the condition for determining whether the thin-walled round sleeve is qualified is: Condition 3: The maximum wall thickness difference d of the thin-walled circular sleeve is less than the threshold th3, which is determined experimentally; Condition 4: There are no burrs on the inner side of the thin-walled round sleeve; If conditions 3 and 4 are met simultaneously, the thin-walled round sleeve is deemed qualified; otherwise, it is deemed unqualified.

Images