A machine vision-based indexable insert edge passivation layer width measurement and analysis method

Abstract

This invention provides a machine vision-based method for measuring the passivation layer width of indexable cutting tools, belonging to the field of machine vision measurement technology. The main steps are as follows: First, a standard gauge block image is acquired using an image acquisition device and marked with pixel equivalents. Then, an image of the cutting edge surface of the indexable cutting tool is acquired at the same object distance. The edge features of the cutting tool image are enhanced through preprocessing. The preprocessed image is then segmented to extract its actual passivation layer contour image. Simultaneously, its ideal passivation layer contour image is obtained by equidistantly offsetting the outer contour inwards. The difference between the actual and ideal passivation layer contour images is calculated, and the passivation layer width of different regions of the cutting tool is obtained by measuring each region. This method can automatically measure the passivation layer width of the cutting tool and reflect the passivation status of different regions of the cutting edge.

Description

Technical Field

[0001] This invention relates to the field of machine vision measurement, and specifically to a method for measuring and analyzing the width of the passivation layer on the cutting edge of a indexable cutting tool based on machine vision. Background Technology

[0002] Modern manufacturing processes involve roughing, finishing, assembly, inspection, and packaging. Cutting tools are indispensable in this industrial chain, their function being to remove excess material from metal parts. In response to the demands for automation and high efficiency in manufacturing, cutting tools have evolved accordingly. To reduce the time spent replacing cutting tools after edge wear, indexable inserts have emerged. These inserts have at least two pre-machined cutting edges, saving time on tool changes and setting, and improving machine tool utilization.

[0003] The manufacturing process of indexable inserts involves grinding, edge treatment, and surface finishing to shape the prepared material into insert products. During grinding, the cutting edge of the insert, after being sharpened with ordinary or diamond grinding wheels, will have varying degrees of micro-serrations and chipping, typically ranging from 0.01-0.05 mm in size, but sometimes exceeding 0.1 mm. In actual use, these micro-defects, if left untreated, can easily lead to chipping and propagation, accelerating wear and even causing the blade to break. Therefore, an edge treatment method is needed to address these defects. Insert passivation treatment removes serrations and chipping, transforming the originally sharp cutting edge into a rounded arc, typically with a diameter of 30-80 micrometers.

[0004] The degree of blade edge passivation directly affects blade performance. Proper passivation treatment can improve cutting stability and enhance surface finish. However, if the passivation exceeds the design range, blade performance will deteriorate. Therefore, precise measurement of the passivation parameters is necessary. Current measurement methods are mainly divided into contact and non-contact methods. Contact methods require manual measurement of the passivation layer width, which is time-consuming, labor-intensive, and inefficient. Therefore, an automated method for measuring the passivation layer width is needed to meet the requirements of large-scale blade passivation inspection. Summary of the Invention

[0005] To address the aforementioned issues, a machine vision-based method for measuring the passivation layer width of indexable cutting tools is proposed. This method can automatically measure the passivation layer width in different regions of the indexable cutting tool. The present invention employs the following technical solution:

[0006] A machine vision-based method for measuring and analyzing the width of the passivation layer of a indexable cutting tool includes the following steps:

[0007] Step S1: Acquire images of standard gauge blocks using an image acquisition device to determine pixel equivalents;

[0008] Step S2: Adjust the camera height to match the object distance in step S1, and acquire an image of the cutting edge of the blade surface to be inspected.

[0009] Step S3: Preprocess the acquired image to remove image noise and enhance image edge features to obtain a preprocessed image;

[0010] Step S4: Perform segmentation processing on the preprocessed blade image to extract the actual passivation layer contour image of the blade;

[0011] Step S5: Perform equidistant contour offset processing on the outer contour of the obtained actual passivation layer contour image of the blade to obtain the ideal passivation layer contour image of the blade.

[0012] Step S6: Difference the actual passivation layer contour image of the blade with the ideal passivation layer contour image of the blade, and measure the obtained area;

[0013] Step S7: Add or subtract the measurement value obtained in step S6 and the offset distance set in step S5, and then obtain the passivation layer width of the region according to the pixel equivalent marked in step S1.

[0014] The specific steps in step S4 include:

[0015] S4-1, On the image preprocessed in step S3, set an inscribed rectangular region with the largest outline;

[0016] S4-2, use a matrix to traverse the largest inscribed rectangle region, then calculate the average value of the pixels in the matrix, and record the coordinates of the median value of the region with the largest and smallest average values;

[0017] S4-3, take this coordinate point as two seed pixels, and randomly select two starting seed pixels within the largest inscribed rectangle area as the seed pixel set based on region segmentation.

[0018] S4-4 uses a region growing algorithm to segment the blade passivation layer contour image.

[0019] The specific steps in step S5 include:

[0020] Step S5-1: Perform image inversion processing on the blade passivation layer contour image;

[0021] Step S5-2: Using the outer contour of the passivation layer as the boundary, set different pixel values ​​for the regions inside and outside the boundary to obtain a binary mask image;

[0022] Step S5-3: Perform morphological dilation on the binary mask image to shrink its outer contour boundary inward. This is equivalent to offsetting the outer contour inward, thus obtaining a binary mask offset image. The offset distance value is consistent with the passivation layer width value of the blade design.

[0023] Step S5-4: Difference is performed between the binary mask image and the binary mask offset image to obtain the region as the ideal passivation layer contour image of the blade.

[0024] The specific steps in step S6 include:

[0025] Step S6-1: Difference is performed between the ideal passivation layer contour image of the blade and the actual passivation layer contour image of the blade. The resulting area is the region where the passivation value of the blade edge is smaller than the design value.

[0026] Step S6-2: Difference is performed between the actual passivation layer contour image of the blade and the ideal passivation layer contour image of the blade. The resulting area is the area where the passivation value of the blade edge is larger than the design value.

[0027] Step S6-3: Measure the acquired target area and determine its pixel width;

[0028] The specific steps in step S7 include:

[0029] S7-1. For areas where the passivation value is smaller than the design value, the width of the passivation layer in that area is obtained by subtracting the measured width value from the specified offset distance in step S5 and then multiplying it by the pixel equivalent.

[0030] S7-2, For areas where the passivation value is larger than the design value, the passivation layer width of that area is obtained by adding the specified offset distance from step S5 to the obtained width value and then multiplying it by the pixel equivalent.

[0031] Beneficial effects of the invention:

[0032] According to the present invention, a machine vision-based method for measuring the passivation layer width of indexable cutting blades is provided for measuring the passivation layer width of indexable cutting blades. This method can measure the passivation layer width of cutting edges with complex shapes and obtain relatively accurate results. Furthermore, it provides corresponding passivation values ​​for different regions of the blade, enabling a comprehensive evaluation of the blade's passivation condition. Attached Figure Description

[0033] Figure 1 This is a schematic diagram of the machine vision-based method for measuring the passivation layer width of a toggleable cutting blade, as described in an embodiment of the present invention.

[0034] Figure 2The image shows the blade edge captured in an embodiment of the present invention.

[0035] Figure 3 This is the actual image of the passivation layer contour extracted in the embodiments of the present invention;

[0036] Figure 4 This is a process diagram of obtaining an ideal image of the passivation layer contour in an embodiment of the present invention. Detailed Implementation

[0037] The following is in conjunction with the appendix Figure 1-4 The specific implementation of the present invention will be illustrated by the examples.

[0038] This embodiment provides a machine vision-based method for measuring the width of the passivation layer of a toggleable cutting tool. Figure 1 This is a flowchart of the method for measuring the passivation layer width of an indexable cutting tool according to an embodiment of the present invention. The following is in conjunction with... Figure 1 This embodiment will be described.

[0039] Step S1: Acquire images of standard gauge blocks using an image acquisition device to determine pixel equivalents;

[0040] In this embodiment, the specific process of pixel calibration is as follows: Let the pixel equivalent be k, the actual width of the standard block be l, and the pixel width of the acquired standard block image be p, then the pixel equivalent is k = l / n.

[0041] Step S2: Adjust the camera height to match the framing distance in step S1, and acquire an image of the surface cutting edge of the blade to be inspected.

[0042] In this embodiment, as Figure 2 As shown, the acquired blade edge surface image is divided into two parts: the striped texture image distributed on the surface and the blunting layer contour image at the edge of the blade.

[0043] Step S3: Preprocess the acquired image to remove noise and enhance image edge features to obtain a preprocessed image.

[0044] In this embodiment, the preprocessing includes using bilateral filtering to remove noise and smooth the image while preserving the edge information of the image. Then, the smoothed image is sharpened using the Laplacian operator to enhance the image's text and edge features.

[0045] Step S4: Perform segmentation processing on the preprocessed blade image to extract the actual passivation layer contour image of the blade;

[0046] Specifically, step S4 includes the following steps:

[0047] S4-1, On the image preprocessed in step S3, set an inscribed rectangular region with the largest outline;

[0048] S4-2, use a matrix to traverse the largest inscribed rectangle region, then calculate the average value of the pixels in the matrix, and record the coordinates of the median value of the region with the largest and smallest average values;

[0049] S4-3, take this coordinate point as two seed pixels, and randomly select two starting seed pixels within the largest inscribed rectangle area as the seed pixel set based on region segmentation.

[0050] S4-4 uses a region growing algorithm to segment the blade passivation layer contour image.

[0051] Specifically, region-growing-based image segmentation algorithms leverage the similarity of background grayscale distribution, grouping pixels with similar properties together to construct segmentation regions. Starting with a set of seed points, neighboring pixels with similar grayscale values ​​are appended to each seed in the growing region, completing image segmentation. For the blade edge surface image in this case, the surface texture portion is relatively large, and the grayscale distribution is also quite consistent; therefore, seed pixels are selected from the blade surface texture image. The blade surface texture consists of two parts: a gray striped texture and a white striped background. Therefore, matrix traversal is used to find the coordinates of the white and black pixels with the largest local average values. Simultaneously, several initial seed pixels are randomly selected within this region to avoid interference from random noise in the image.

[0052] Figure 3 The image shown is the extracted passivation layer contour image in an embodiment of the present invention. Figure 3 As shown, the passivation layer image of the blade edge was completely extracted.

[0053] Step S5: Perform a fixed-distance contour offset on the outer contour of the obtained actual passivation layer contour image of the blade to obtain the ideal passivation layer contour image of the blade.

[0054] Figure 4 This is a process diagram for obtaining an ideal image of the passivation layer contour in an embodiment of the present invention. Specifically, step S5 includes the following steps:

[0055] Step S5-1: Perform image inversion processing on the blade passivation layer contour image;

[0056] Step S5-2: Using the outer contour of the passivation layer as the boundary, set different pixel values ​​for the regions inside and outside the boundary to obtain a binary mask image;

[0057] Step S5-3: Perform morphological dilation on the binary mask image to move its outer contour boundary inward. This is equivalent to offsetting the outer contour inward. The offset distance value is consistent with the passivation layer width value of the blade design, thereby obtaining a binary mask offset image.

[0058] Step S5-4: Difference is performed between the binary mask image and the binary mask offset image to obtain the region as the ideal passivation layer contour image of the blade.

[0059] Specifically, in the binary mask image obtained in step S5-2, it can be observed that the passivation layer image, i.e., the white area, has inconsistent width. During the passivation process of the blade edge, it is necessary to ensure that the passivation layer width is within the designed reasonable range, so the passivation layer width needs to be measured. This step obtains the ideal passivation layer contour image of the blade, i.e., the passivation layer width of this image is equal to the designed passivation layer width. Using this as a template, by comparison, the areas where the actual passivation layer width of the blade deviates from the ideal passivation layer width can be identified.

[0060] Step S6: Difference the actual passivation layer contour image of the blade with the ideal passivation layer contour image of the blade, and measure the obtained area;

[0061] Specifically, step S6 includes the following steps:

[0062] S6-1, the ideal passivation layer contour image of the blade is compared with the actual passivation layer contour image of the blade, and the resulting area is the area where the blade edge passivation value is smaller than the design value;

[0063] Specifically, if the passivation layer width in a certain area is smaller than the design value, the area obtained by subtracting the actual passivation layer contour image of the blade from the ideal passivation layer contour image is the area where the passivation value is smaller than the design value.

[0064] S6-2, the actual passivation layer contour image of the blade is compared with the ideal passivation layer contour image of the blade, and the resulting area is the area where the passivation value of the blade edge is larger than the design value;

[0065] Specifically, if the passivation width of a certain area is larger than the design value, the area obtained by subtracting the ideal passivation layer contour image of the blade from the actual passivation layer contour image is the area where the passivation value is larger than the design value.

[0066] S6-3, Measure the acquired target area and determine its width;

[0067] Specifically, the acquired region is a bar shape, and the maximum pixel width of the bar is calculated.

[0068] Step S7: Add or subtract the measurement value obtained in step S6 and the offset distance set in step S5, and then obtain the passivation layer width of the region according to the pixel equivalent marked in step S1.

[0069] Specifically, step S7 includes the following steps:

[0070] S7-1. For areas where the passivation value is smaller than the design value, the width of the passivation layer in that area is obtained by subtracting the measured width value from the specified offset distance in step S5 and then multiplying it by the pixel equivalent.

[0071] S7-2, For areas where the passivation value is larger than the design value, the passivation layer width of that area is obtained by adding the specified offset distance from step S5 to the obtained width value and then multiplying it by the pixel equivalent.

[0072] The main innovation of this invention is:

[0073] 1) When obtaining the ideal contour image of the blade passivation layer, a morphology-based dilation operation was creatively used to perform equidistant offset processing on the outer contour of the blade passivation layer, thereby obtaining the ideal passivation layer image of the blade.

[0074] 2) In measuring the width of the passivation layer of the cutting edge, an image difference-based method was creatively used to measure the width of the passivation layer in different areas of the cutting edge.

[0075] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely preferred examples and are not intended to limit the invention. Various changes and modifications can be made to the present invention without departing from its novel spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for measuring and analyzing the width of the passivation layer of a indexable cutting tool based on machine vision, characterized in that, Including the following: Step S1: Acquire images of standard gauge blocks using an image acquisition device to determine pixel equivalents; Step S2: Adjust the camera height to match the object distance in step S1, and acquire an image of the cutting edge of the blade surface to be inspected. Step S3: Preprocess the acquired image to remove image noise and enhance image edge features to obtain a preprocessed image; Step S4: Perform segmentation processing on the preprocessed blade image to extract the actual passivation layer contour image of the blade; Step S5: Perform equidistant contour offset processing on the outer contour of the obtained actual passivation layer contour image of the blade to obtain the ideal passivation layer contour image of the blade. Step S6: Difference the actual passivation layer contour image of the blade with the ideal passivation layer contour image of the blade, and measure the obtained area; In step S7, the measurement value obtained in step S6 and the offset distance set in step S5 are added or subtracted, and the passivation layer width of the region can be obtained according to the pixel equivalent marked in step S1.

2. The method for measuring and analyzing the passivation of indexable cutting tools based on machine vision according to claim 1, characterized in that, Step S4 includes the following steps: Step S4-1: On the image preprocessed in step S3, define a rectangle region inscribed in the contour. Step S4-2: Use a matrix to traverse the inscribed rectangular region, then calculate the average value of the pixels in the matrix, and record the coordinates of the median value of the region with the maximum and minimum average values. Step S4-3: Use this coordinate point as two seed pixel points, and randomly select two starting seed pixel points within the largest inscribed rectangle area as the seed pixel point set based on region segmentation. Step S4-4: Use the region growing algorithm to extract the actual passivation layer contour image of the blade.

3. The machine vision-based method for measuring and analyzing the passivation of indexable cutting tools according to claim 1, characterized in that, Step S5 includes the following steps: Step S5-1: Invert the actual passivation layer contour image of the blade. Step S5-2: Using the outer contour of the passivation layer as the boundary, set different pixel values ​​for the regions inside and outside the boundary to obtain a binary mask image; Step S5-3: Perform morphological dilation on the binary mask image to move its outer contour boundary inward. This is equivalent to offsetting the outer contour inward. The offset distance value is consistent with the passivation layer width value of the blade design. Finally, a binary mask offset image is obtained. Step S5-4: Difference is performed between the binary mask image and the binary mask offset image to obtain the region as the ideal passivation layer contour image of the blade.

4. The method for measuring and analyzing the passivation of indexable cutting tools based on machine vision according to claim 1, characterized in that, Step S6 includes the following steps: Step S6-1: Difference is performed between the ideal passivation layer contour image of the blade and the actual passivation layer contour image of the blade. The resulting area is the region where the passivation value of the blade edge is smaller than the design value. Step S6-2: Difference is performed between the actual passivation layer contour image of the blade and the ideal passivation layer contour image of the blade. The resulting area is the area where the passivation value of the blade edge is larger than the design value. Step S6-3: Measure the acquired target area and determine its pixel width.

5. The machine vision-based method for measuring and analyzing the passivation of indexable cutting tools according to claim 1, characterized in that, Step S7 includes the following steps: Step S7-1: For areas where the passivation value is smaller than the design value, subtract the measured width value from the specified offset distance in step S5, and then multiply by the pixel equivalent to obtain the passivation layer width of that area. Step S7-2: For areas where the passivation value is larger than the design value, add the obtained width value to the specified offset distance in step S5, and then multiply by the pixel equivalent to obtain the passivation layer width for that area.

Images