A method, device and medium for identifying thickness of coated channel steel

By performing specific processing and contour detection on the end face image of the composite steel cladding channel steel, the problem that the existing technology cannot accurately calculate the thickness of the cladding is solved, efficient and accurate thickness recognition is achieved, and detection efficiency and accuracy are improved.

CN118129617BActive Publication Date: 2024-12-06CENT SOUTH UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410249598.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-03-05
Publication Date
2024-12-06
Estimated Expiration
2044-03-05

AI Technical Summary

Technical Problem

The prior art cannot efficiently and accurately calculate the thickness of composite steel cladding, resulting in easy corrosion and affecting service life.

Method used

By acquiring the end surface image of the cladding channel steel, performing grayscale processing, noise reduction processing and adaptive threshold processing, rotating the image to level the bottom surface of the channel steel, detecting the center and contour, dividing the area for threshold processing, and calculating the cladding thickness.

Benefits of technology

It realizes high robustness and accurate identification of the thickness of the cladding channel steel under uneven lighting conditions, high detection efficiency, simple mathematical model, and um-level recognition accuracy, and short output time as a result.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118129617B_ABST
    Figure CN118129617B_ABST
Patent Text Reader

Abstract

The present invention discloses a method, device and medium for identifying the thickness of coated channel steel, which is used to solve the problem that the existing technology cannot efficiently and accurately calculate the coating thickness of coated steel. The method of the present invention is: obtain the end face image of the coated channel steel to be detected, grayscale it, reduce noise and perform adaptive threshold processing on it, and rotate the obtained image until the bottom surface of the channel steel is horizontal; by sequentially performing edge detection and center detection on the rotated image, the center position of the turning arc on both sides of the groove of the channel steel is obtained; then based on the two obtained center positions, the image is divided into a left rectangular area, a middle rectangular area and a right rectangular area; then the left and right rectangular areas are subjected to adaptive threshold processing, and the middle rectangular area is directly subjected to threshold processing; contour detection is performed on the image after the regional threshold processing to obtain the outer contour and inner contour of the coated channel steel; according to the outer contour and inner contour of the coated channel steel, the coating thickness of the coated channel steel is calculated.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention belongs to the field of coating thickness detection, and is characterized by a coating channel steel thickness identification method, device and medium. Background Art

[0002] Composite steel is a new type of steel that uses low-alloy steel as the base material and stainless steel as the cladding. The corrosion resistance of this steel is close to that of stainless steel, but the cost is low, and it can be widely used in roads, bridges and the power industry. The China Steel Industry Development Trend Analysis and Future Investment Forecast Report pointed out that in 2023, my country's demand for corrosion-resistant steel will reach one million tons, and this demand will continue to grow in the next five years.

[0003] It is found in actual use that corrosion will preferentially occur at the thinnest part of the composite steel coating. Once corrosion occurs, the corrosion resistance of the composite steel and the composite surface bonding performance will be greatly reduced, thus affecting the service life of the coated steel. Therefore, the detection of composite steel coating thickness is imminent.

[0004] Chinese patent CN202011495555.X discloses a method for identifying the thickness of stainless steel coated steel bars. Since steel bars are rods, this method is to first find the center of the steel bar and then draw a straight line through the center of the circle to obtain the corresponding steel bar coating thickness. However, the detection object this time is steel sections, which are not rods, and the coating thickness cannot be obtained by finding the center of the circle and drawing a straight line.

[0005] Chinese patent CN202210250841.2 discloses a method for identifying the coating thickness and evaluating the thickness uniformity of composite steel bars. This method also finds the center of the steel bar and then calculates the coating thickness through the minimum circle radius. Due to the different detection objects, this method is still not applicable to composite steel sections.

[0006] Chinese patent CN201921320495.0 discloses a coating thickness detector, which measures based on ultrasonic reflection. However, the composite steel base and coating are both made of steel. When ultrasonic waves pass through two metals, the reflected signal is not obvious.

[0007] Chinese patent CN202210188561.3 discloses a non-contact method for measuring the thickness of a non-conductive coating on a metal surface. This method mainly measures the non-conductive coating, but the composite steel coating is stainless steel, which is a conductive material.

[0008] In summary, the existing detection technologies are unable to efficiently and accurately calculate the thickness of composite steel coatings. Summary of the invention

[0009] The present invention provides a method, device and medium for identifying the thickness of a clad channel steel, which can solve the problem in the prior art that the cladding thickness of the clad steel cannot be calculated efficiently and accurately.

[0010] In order to achieve the above technical objectives, the present invention adopts the following technical solutions:

[0011] On the one hand, a method for identifying the thickness of a cladding channel steel is provided, comprising:

[0012] S1, obtaining an end face image of the coated channel steel to be inspected;

[0013] S2, graying, denoising and adaptive threshold processing are performed on the end surface image of the cladding channel steel;

[0014] S3, rotating the image obtained after processing in step S2 until the bottom surface of the channel steel is horizontal;

[0015] S4, by sequentially performing edge detection and center detection on the image rotated in step S3, the center positions of the turning arcs on both sides of the groove of the channel steel are obtained; then based on the two center positions obtained, the image is divided into a left rectangular area, a middle rectangular area and a right rectangular area; then adaptive threshold processing is performed on the left and right rectangular areas, and threshold processing is directly performed on the middle rectangular area;

[0016] S5, performing contour detection on the image after the regional threshold processing in step S4 to obtain the outer contour and inner contour of the cladding channel steel;

[0017] S6, calculate the coating thickness of the coated channel steel according to the outer contour and inner contour of the coated channel steel.

[0018] Furthermore, in step S1, an industrial camera is used to obtain an end face image of the clad channel steel with a black background.

[0019] Furthermore, the adaptive threshold processing in step S2 is used to segment the end face of the channel steel from the background, and to initially segment the cladding area and the steel base area in the end face of the channel steel.

[0020] Further, step S3 rotates the image until the bottom surface of the channel steel is horizontal, specifically:

[0021] First, edge contour detection is performed on the image obtained after processing in step S2;

[0022] Then find the vertices A1 and A2 on the left and right sides of the edge contour, and set their coordinates to

[0023] Next, calculate the horizontal angle α of the straight line A1A2 connecting the two points. The formula is as follows:

[0024]

[0025] Finally, the image is rotated clockwise around point A1 by an angle α.

[0026] Furthermore, when performing circle center detection in step S4, the radius range of the circle is set to [R-δ1, R+δ2]; wherein R is the arc radius at the turning points on both sides of the groove of the channel steel, and δ2 and δ1 are the upper and lower thresholds of the circle center detection radius, respectively.

[0027] Further, step S6 calculates the thickness of each point of the inner contour of the coating; wherein, any point N of the inner contour i Thickness F i The calculation methods include:

[0028] S6.1, get point N i N adjacent points on the inner contour in the clockwise direction i+1 , suppose two adjacent points N i 、N i+1 The coordinates of

[0029] S6.2, calculate the straight line N connecting two points i N i+1 Horizontal angle

[0030] S6.3, calculate the coating at the inner contour point N according to the horizontal angle i Thickness at:

[0031] (1) When the horizontal angle β i = 0°, passing through point N i Draw a vertical line that intersects the outer contour at two points: and Calculate W i1 Dot and W i2 The coating thickness corresponding to the point The formula is as follows:

[0032]

[0033]

[0034] Where k is the actual physical size of each pixel;

[0035] Compare and The smaller value of the two is N i The corresponding coating thickness F i ;

[0036] (2) When the horizontal angle β i ∈(0°, 90°), passing through point Ni Draw a vertical line to N i N i+1 The straight line has two intersection points with the outer contour, which are and Calculate Q i1 Dot and Q i2 The coating thickness corresponding to the point The formula is as follows:

[0037]

[0038]

[0039] Compare and The smaller value of the two is N i The corresponding coating thickness F i ;

[0040] (3) When the horizontal angle β i =90°, passing through point N i Draw a horizontal line, which also has two intersection points with the outer contour. and Then measure T i1 Dot and T i2 The coating thickness corresponding to the point The formula is as follows:

[0041]

[0042]

[0043] Compare and The smaller value of the two is N i The corresponding coating thickness F i .

[0044] Furthermore, according to the thickness at each point of the inner contour of the coating, the average value, minimum value and maximum value of the thickness of the coating at all points of the inner contour are statistically calculated.

[0045] A device for identifying the thickness of a cladding channel steel, comprising:

[0046] An image acquisition module is used to: acquire an end face image of the cladding channel steel to be inspected;

[0047] The pre-processing module is used to: grayscale, denoise and adaptively threshold the end surface image of the cladding channel steel;

[0048] The image rotation module is used to: rotate the image obtained by the preprocessing module until the bottom surface of the channel steel is horizontal;

[0049] The regional threshold processing module is used to: firstly perform edge detection and circle center detection on the rotated image in sequence to obtain the center positions of the turning arcs on both sides of the groove of the channel steel; then, based on the two obtained center positions, divide the image into a left rectangular area, a middle rectangular area and a right rectangular area; then, perform adaptive threshold processing on the left and right rectangular areas, and directly perform threshold processing on the middle rectangular area;

[0050] The contour extraction module is used to: perform contour detection on the image after the regional threshold processing to obtain the outer contour and inner contour of the cladding channel steel;

[0051] The coating thickness calculation module is used to calculate the coating thickness of the coating channel steel according to the outer contour and inner contour of the coating channel steel.

[0052] A computer-readable storage medium stores a computer program, wherein the computer program, when executed by a processor, implements any of the above-mentioned methods for identifying the thickness of a coated channel steel.

[0053] Compared with the prior art, the thickness identification method, device and medium of the coated channel steel provided by the present invention have the following beneficial technical effects:

[0054] 1) The recognition method system has high robustness and can effectively identify the coating thickness of the coated channel steel under uneven lighting conditions.

[0055] 2) The mathematical model of this recognition method is simple and reliable, and its recognition accuracy can reach um level;

[0056] 3) This recognition method has high detection efficiency, and it only takes 100ms from model calculation to result output. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 is the end face image of the clad channel steel after cutting obtained in step S1 of the embodiment of the present invention;

[0058] Figure 2 is the end face image obtained after processing in step S2 of the embodiment of the present invention;

[0059] Figure 3 is a schematic diagram of obtaining the horizontal angle α in step S3 of an embodiment of the present invention;

[0060] Figure 4 is the adaptive threshold image after rotation in step S3 of the embodiment of the present invention;

[0061] Figure 5 is the end face image of the channel steel after a portion of the outline is deleted in step S4 of the embodiment of the present invention;

[0062] Figure 6is the center point detection result of the channel steel end face image in step S4 of the embodiment of the present invention;

[0063] Figure 7 is a schematic diagram of a rectangular area obtained by dividing the area in step S4 of an embodiment of the present invention;

[0064] Figure 8 is the result obtained by the regional threshold processing in step S4 of the embodiment of the present invention;

[0065] Fig. 9 is a schematic diagram of the inner and outer contours obtained in step S5 of the embodiment of the present invention;

[0066] Fig.10 It is a schematic diagram of obtaining the coating thickness when the horizontal angle is 0° in step S6 of an embodiment of the present invention;

[0067] Fig.11 It is a schematic diagram of obtaining the coating thickness when the horizontal angle is (0°, 90°) in step S6 of an embodiment of the present invention. DETAILED DESCRIPTION

[0068] The following is a detailed description of an embodiment of the present invention. This embodiment is based on the technical solution of the present invention, and provides a detailed implementation method and a specific operation process to further explain the technical solution of the present invention.

[0069] One embodiment of the present invention provides a method for identifying the thickness of a stainless steel coated steel bar, comprising:

[0070] S1. Cut a 10 mm thick coated channel steel sample using a wire cutting machine, and then use an industrial camera to take an image of the end face of the channel steel. To improve the image contrast, the background is black. The end face image of the coated channel steel to be tested obtained in this embodiment is as follows: Figure 1 shown.

[0071] S2, respectively use cvtColor function, GaussianBlur image and adaptiveThreshold function to grayscale, Gaussian noise reduction and adaptive threshold processing on the end face image of the cladding channel steel; the end face image obtained in this embodiment is as follows Figure 2 shown.

[0072] S3. Use the Canny function to detect the edge contour of the image after adaptive threshold processing, and then find the vertices A1 and A2 on the left and right sides. In the contour coordinates, the point with the smallest horizontal and vertical coordinates is the left vertex, and the point with the largest horizontal coordinate and the smallest vertical coordinate is the right vertex. Figure 3 As shown, the coordinates of the left and right vertices are specifically expressed in and Indicates that the horizontal angle α of the straight line A1A2 connecting the two points is calculated, and the formula is as follows:

[0073]

[0074] Finally, the image is rotated clockwise around point A1 by an angle α. The rotated image obtained in this embodiment is as follows Figure 4 shown.

[0075] S4. Since the channel steel is too large, there is uneven illumination during image acquisition. The adaptive threshold processing of the entire channel steel end face image does not produce a good effect. This embodiment adopts regional threshold processing. The specific steps are as follows:

[0076] S4-1. Use the Canny function to perform edge detection on the rotated adaptive threshold image.

[0077] S4-2, detect the length of each contour, and delete the contour whose length is less than the preset value. Figure 5 Image shown.

[0078] S4-3. Use the Hough function to detect the center of the channel steel image after deleting the contour. Specifically, the radius range of the center detection is set to [R-δ1, R+δ2] according to the arc radius R at the turning points on both sides of the channel steel. δ2 and δ1 are the upper and lower thresholds of the center detection radius. In this embodiment, the radius range of the center detection is set to (100, 200), and the detected center coordinates are set to O1 (x3, y3) and O2 (x4, y4), respectively. Figure 6 shown.

[0079] S4-4, according to the detected center coordinates, the channel steel image is divided into a left rectangular area, a middle rectangular area and a right rectangular area, such as Figure 7 As shown, the vertex coordinates of the three rectangular areas are as follows: left rectangle 1 [(0,0), (x3,0), (x3,l2), (0,l2)], middle rectangle 2 [(x3,0), (x4,0), (x4,l2), (x3,l2)], right rectangle 3 [(x4,0), (l1,0), (l1,l2), (x4,l2)], where l1 is the total number of pixels in the horizontal direction of the image, and l2 is the total number of pixels in the vertical direction of the image.

[0080] S4-5, use the adaptiveThreshold function to perform adaptive threshold processing on the left rectangular area and the right rectangular area, and use the threshold function to perform direct threshold processing on the middle rectangular area, where the threshold of the direct threshold processing is 170. The results of the regional threshold processing are as follows: Figure 8 shown.

[0081] S5. Use the Canny function to perform contour detection on the image after threshold processing in each region, where the outer contour is called the outer contour and the inner contour is called the inner contour. Fig. 9 shown.

[0082] S6. Calculate the coating thickness of the coated channel steel.

[0083] In step S6, the thickness at each point of the inner contour of the coating is first calculated, and then the average value, minimum value and maximum value of the thickness of the coating at all points of the inner contour are counted based on the thickness at each point of the inner contour of the coating.

[0084] Any point N in the inner contour i Thickness F i The calculation methods include:

[0085] S6.1, get point N i N adjacent points on the inner contour in the clockwise direction i+1 , suppose two adjacent points N i 、N i+1 The coordinates of

[0086] S6.2, calculate the straight line N connecting two points i N i+1 Horizontal angle

[0087] S6.3, calculate the coating at the inner contour point N according to the horizontal angle i Thickness at:

[0088] (1) When the horizontal angle β i = 0°, if Fig.10 As shown, through point N i Draw a vertical line that intersects the outer contour at two points: and Calculate W i1 Dot and W i2 The coating thickness F corresponding to the point Wi1 、F Wi2 , the formula is as follows:

[0089]

[0090]

[0091] Where k is the actual physical size of each pixel;

[0092] Comparison F Wi1 and F Wi2 The smaller value of the two is N i The corresponding coating thickness F i ;

[0093] (2) When the horizontal angle β i ∈(0°, 90°), such as Fig.11 As shown, through point N i Draw a vertical line to N i N i+1 The straight line, The straight line has two intersection points with the outer contour, which are and Calculate Q i1 Dot and Q i2 The coating thickness corresponding to the point The formula is as follows:

[0094]

[0095]

[0096] Compare and The smaller value of the two is N i The corresponding coating thickness F i ;

[0097] (3) When the horizontal angle β i =90°, passing through point N i Draw a horizontal line, which also has two intersection points with the outer contour. and Then measure T i1 Dot and T i2 The coating thickness corresponding to the point The formula is as follows:

[0098]

[0099]

[0100] Comparison F Ti1 and F Ti2 The smaller value of the two is N i The corresponding coating thickness F i .

[0101] Statistical thickness average F of the coating at all n points of the inner contour 平均 , minimum value F 最小 and the maximum value F 最大 :

[0102] F 平均 =(F1+F2+…F n ) / n

[0103] F 最大=max(F1,F2,…F n )

[0104] F 最小 =min(F1,F2,…F n )

[0105] In the example of the present invention, the average value, maximum value and minimum value of the coating thickness of the coated channel steel obtained by the above method are 0.855 mm, 2.054 mm and 0.456 mm respectively.

[0106] The above embodiments are preferred embodiments of the present application. Ordinary technicians in this field can also make various changes or improvements on this basis. Without departing from the overall concept of the present application, these changes or improvements should fall within the scope of protection required by the present application.

Claims

1. A method for identifying the thickness of a coated channel steel, characterized in that: include: S1, obtaining an end face image of the coated channel steel to be inspected; S2, graying, denoising and adaptive threshold processing are performed on the end surface image of the cladding channel steel; S3, rotating the image obtained after processing in step S2 until the bottom surface of the channel steel is horizontal; S4, by sequentially performing edge detection and center detection on the image rotated in step S3, the center positions of the turning arcs on both sides of the groove of the channel steel are obtained; then based on the two center positions obtained, the image is divided into a left rectangular area, a middle rectangular area and a right rectangular area; then adaptive threshold processing is performed on the left and right rectangular areas, and threshold processing is directly performed on the middle rectangular area; S5, performing contour detection on the image after the regional threshold processing in step S4 to obtain the outer contour and inner contour of the cladding channel steel; S6, calculate the coating thickness of the coated channel steel according to the outer contour and inner contour of the coated channel steel.

2. The method for identifying the thickness of a coated channel steel according to claim 1, characterized in that: Step S1 uses an industrial camera to obtain an end face image of a clad channel steel with a black background.

3. The method for identifying the thickness of a coated channel steel according to claim 1, characterized in that: The adaptive threshold processing in step S2 is used to segment the end face of the channel steel from the background, and to initially segment the cladding area and the steel base area in the end face of the channel steel.

4. The method for identifying the thickness of a cladding channel steel according to claim 1, characterized in that: Step S3 rotates the image until the bottom surface of the channel steel is horizontal, specifically: First, edge contour detection is performed on the image obtained after processing in step S2; Then find the vertices A1 and A2 on the left and right sides of the edge contour, and set their coordinates to Next, calculate the horizontal angle α of the straight line A1A2 connecting the two points. The formula is as follows: Finally, the image is rotated clockwise around point A1 by an angle α.

5. The method for identifying the thickness of a cladding channel steel according to claim 1, characterized in that: When performing circle center detection in step S4, the radius range of the circle is set to [R-δ1, R+δ2]; wherein R is the arc radius at the turning points on both sides of the groove of the channel steel, and δ2 and δ1 are the upper and lower thresholds of the circle center detection radius, respectively.

6. The method for identifying the thickness of a cladding channel steel according to claim 1, characterized in that: Step S6 calculates the thickness of each point of the inner contour of the coating; wherein, any point N of the inner contour i Thickness F i The calculation methods include: S6.1, get point N i N adjacent points on the inner contour in the clockwise direction i+1 , suppose two adjacent points N i 、N i+1 The coordinates of S6.2, calculate the straight line N connecting two points i N i+1 Horizontal angle S6.3, calculate the coating at the inner contour point N according to the horizontal angle i Thickness at: (1) When the horizontal angle β i = 0°, passing through point N i Draw a vertical line that intersects the outer contour at two points: and Calculate W i1 Dot and W i2 The coating thickness corresponding to the point The formula is as follows: Where k is the actual physical size of each pixel; Compare and The smaller value of the two is N i The corresponding coating thickness F i ; (2) When the horizontal angle β i ∈(0°, 90°), passing through point N i Draw a vertical line to N i N i+1 The straight line has two intersection points with the outer contour, which are and Calculate Q i1 Dot and Q i2 The coating thickness corresponding to the point The formula is as follows: Compare and The smaller value of the two is N i The corresponding coating thickness F i ; (3) When the horizontal angle β i =90°, passing through point N i Draw a horizontal line, which also has two intersection points with the outer contour. and Then measure T i1 Dot and T i2 The coating thickness corresponding to the point The formula is as follows: Compare and The smaller value of the two is N i The corresponding coating thickness F i .

7. The method for identifying the thickness of a cladding channel steel according to claim 6, characterized in that: According to the thickness at each point of the inner contour of the coating, the average, minimum and maximum values ​​of the thickness of the coating at all points of the inner contour are calculated.

8. A device for identifying the thickness of a cladding channel steel, characterized in that: include: An image acquisition module is used to: acquire an end face image of the cladding channel steel to be inspected; The pre-processing module is used to: grayscale, denoise and adaptively threshold the end surface image of the cladding channel steel; The image rotation module is used to: rotate the image obtained by the preprocessing module until the bottom surface of the channel steel is horizontal; The regional threshold processing module is used to: firstly perform edge detection and circle center detection on the rotated image in sequence to obtain the center positions of the turning arcs on both sides of the groove of the channel steel; then, based on the two obtained center positions, divide the image into a left rectangular area, a middle rectangular area and a right rectangular area; then, perform adaptive threshold processing on the left and right rectangular areas, and directly perform threshold processing on the middle rectangular area; The contour extraction module is used to: perform contour detection on the image after the regional threshold processing to obtain the outer contour and inner contour of the cladding channel steel; The coating thickness calculation module is used to calculate the coating thickness of the coating channel steel according to the outer contour and inner contour of the coating channel steel.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for identifying the thickness of the clad channel steel according to any one of claims 1 to 7 is implemented.