A method and system for machining and detecting parts by a milling machine

Abstract

The present invention relates to the field of part detection, and particularly to a method and system for detecting the machining of parts by a milling machine. The method includes: acquiring a color image of a part machined by a milling machine, and converting the color image to obtain a grayscale image; calculating the grayscale anomaly index of pixel points in the grayscale image; calculating the brightness anomaly index of pixel points in the color image; constructing a feature vector of corresponding pixel points by using the grayscale anomaly index and the brightness anomaly index; calculating the scratch index of pixel points, taking the pixel points with the scratch index greater than a preset scratch threshold as marked points, clustering the marked points to obtain a plurality of clustering clusters, and in response to the number of data points in the clustering cluster being greater than a preset number threshold, sending a prompt that there are scratches on the surface of the part. The present invention improves the accuracy of detecting parts machined by a milling machine.

Description

Technical Field

[0001] The present invention relates to the field of part detection, and particularly to a method and system for detecting the machining of parts by a milling machine. Background Art

[0002] A milling machine is a common machining equipment. It is mainly used for machining metal materials, especially in surface machining, profiling machining, inclined plane machining and curve machining. The common machining methods of a milling machine are: face milling, slot milling, side milling, profiling milling and gear milling. Among them, face milling is used to machine the plane of a workpiece, and a face milling cutter is usually used for cutting. During face milling, due to improper tool selection, inappropriate cutting parameters, or insecure fixation of the workpiece or fixture, vibrations of the machine tool, tool or workpiece may occur, further resulting in an uneven machining surface with tool marks, i.e., scratches. Based on the characteristics of milling machining, scratches generally appear on the surface of parts in an approximately straight line.

[0003] The Chinese patent application document with the publication number CN115760696A discloses a surface scratch detection method, which relates to the field of machine vision defect detection technology. The method includes the following steps: using an image acquisition device to acquire the surface image of a product, and performing image processing on the acquired surface image to obtain a region to be detected; preprocessing the image of the region to be detected; performing edge detection on the preprocessed image of the region to be detected, and fusing the obtained different edge detection image results to obtain a fused edge detection image; processing the fused edge detection image to obtain an image with marked edges; and screening scratches on the image with marked edges to obtain the surface scratch detection result.

[0004] The above solution obtains edges by performing edge detection on an image, and then obtains the detection result of scratches on the product surface. However, based on the characteristics of milling machining, the surface of parts machined by a milling machine usually has slight scratches with a shallow scratch depth and is not easy to identify. Using edge detection algorithms is prone to false negatives and false positives, resulting in low accuracy of part detection. Summary of the Invention

[0005] In order to solve the problem that existing scratches on the surface of parts machined by a milling machine are prone to false negatives during detection, resulting in low detection accuracy, the present invention provides a method and system for detecting the machining of parts by a milling machine.

[0006] In a first aspect, the present invention provides a method for detecting the machining of parts by a milling machine, adopting the following technical solution:

[0007] Obtain a color image of a part machined by a milling machine, and convert the color image into a grayscale image;

[0008] Calculate the gray anomaly index of the pixel points in the grayscale image. The gray anomaly index is positively correlated with the absolute value of the difference between the gray value of the corresponding pixel point and the gray values of other pixel points. Calculate the lightness anomaly index of the pixel points in the color image. The lightness anomaly index is positively correlated with the absolute value of the difference between the lightness value of the corresponding pixel point and the lightness values of other pixel points. Construct the feature vector of the corresponding pixel points using the gray anomaly index and the lightness anomaly index. Calculate the scratch index of the pixel points. Take the pixel points with a scratch index greater than the preset scratch threshold as marked points, and cluster the marked points to obtain multiple clusters. In response to the number of data points in the cluster being greater than the preset number threshold, issue a prompt indicating that there are scratches on the part surface.

[0009] Among them, the expression of the scratch index is:

[0010]

[0011] In the formula, represents the scratch index of the m-th pixel point in the grayscale image, represents the modulus of the feature vector of the m-th pixel point, represents the modulus of the feature vector of the n-th neighboring pixel point of the m-th pixel point, and k represents the total number of neighboring pixel points of the m-th pixel point.

[0012] Construct the feature vector using the lightness anomaly index and the gray anomaly index, and then further calculate the scratch index of the pixel points. The scratch index can be used to determine whether the corresponding pixel point area is a scratch area, which is convenient for detecting slight scratches on the part surface, improves the accuracy of the detection of parts processed by the milling machine, and avoids the phenomenon of missed detection of scratch areas.

[0013] Preferably, the calculation method of the gray anomaly index is: perform edge detection on the grayscale image to obtain edge pixel points. The expression of the gray anomaly index is:

[0014]

[0015] In the formula, represents the gray anomaly index of the m-th pixel point in the grayscale image, represents the gray value of the m-th pixel point in the grayscale image, represents the average gray value of the remaining pixel points after removing the edge pixel points in the grayscale image, represents the maximum gray value of the neighboring pixel points of the m-th pixel point in the grayscale image. a represents the number of neighboring pixel points of the m-th pixel point whose gray value is greater than the gray value of the m-th pixel point, and norm represents the normalization function.

[0016] By comparing the gray value of a pixel with the gray values of the remaining pixels, a gray anomaly index is obtained. Using the gray index, the possibility that the corresponding pixel area is a pixel in the scratch area can be initially understood.

[0017] Preferably, the method further includes:

[0018] Construct a first window and a second window respectively centered on the corresponding pixel. The area of the second window is larger than the area of the first window;

[0019] Construct a first gray-level co-occurrence matrix of the first window, and calculate the first contrast of the first gray-level co-occurrence matrix using the contrast calculation formula; construct a second gray-level co-occurrence matrix of the second window, and calculate the second contrast of the second gray-level co-occurrence matrix using the contrast calculation formula; take the ratio of the first contrast to the sum of the first contrast and the second contrast as the weight factor of the gray anomaly index of the corresponding pixel.

[0020] Adjusting the influence degree of the weight factor on the gray anomaly index improves the accuracy of the feature vector.

[0021] Preferably, the expression of the feature vector is:

[0022]

[0023] In the formula, represents the feature vector of the m-th pixel, represents the gray anomaly index of the m-th pixel in the gray image, represents the brightness anomaly index of the m-th pixel in the gray image, and β represents the weight factor of the gray anomaly index of the preset m-th pixel.

[0024] Construct a feature vector using the gray anomaly index and the brightness anomaly index, so that the possibility that the corresponding pixel is a pixel in the scratch area can be further reflected through the feature vector.

[0025] Preferably, the calculation method of the scratch threshold is: set the scratch threshold as H, obtain multiple marked points, select any one marked point as the center point, and connect the center point with the marked points to obtain multiple connecting lines;

[0026] Construct an objective function, and the expression is:

[0027]

[0028] In the formula, represents the objective function regarding the scratch threshold H, represents the variance of the cosine values of the angles between multiple connecting lines and a preset reference line, Entropy represents the gray value of the pixel corresponding to the marked point, and b represents the number of connection lines; the numerical value of H is obtained by solving the minimum value of the objective function; multiple marked points are repeatedly selected as the center points, and further multiple numerical values of H are obtained, and the average value of the obtained numerical values of H is used as the scratch threshold.

[0029] Construct the objective function from parameters in multiple dimensions, and the corresponding scratch area can be accurately obtained by using the objective function, providing a theoretical basis for screening the pixel points in the scratch area.

[0030] Preferably, the method for selecting the reference line is: select the line in any one of the directions of 0°, 45°, and 90° as the reference line.

[0031] Preferably, the minimum value of the objective function is solved by using the swarm intelligence optimization algorithm.

[0032] Preferably, the method further includes: fitting the data points in the largest clustering cluster by using the least square method to obtain a fitting line, and further obtaining the scratch area.

[0033] Preferably, the method further includes: performing edge detection on the grayscale image by using the canny algorithm to obtain edge pixel points.

[0034] In a second aspect, the present invention provides a part milling machine processing detection system, adopting the following technical solutions:

[0035] A part milling machine processing detection system includes: a processor and a memory, and the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the above-mentioned part milling machine processing detection method is implemented.

[0036] Generate a computer program for the above-mentioned part milling machine processing detection method and store it in the memory to be loaded and executed by the processor. Thus, a system is made according to the memory and the processor, which is convenient to use.

[0037] The present invention has the following technical effects:

[0038] 1. Construct a feature vector by using the brightness anomaly index and the gray anomaly index, and further calculate the scratch index of the pixel point. The scratch index can be used to judge whether the corresponding pixel point area is a scratch area, so as to facilitate the detection of slight scratches on the part surface, improve the accuracy of the detection of parts processed by the milling machine, and avoid the phenomenon of missed detection of the scratch area.

[0039] 2. Adaptively calculate the weight factor of the gray anomaly index for each pixel point, and use the weight factor to adjust the influence degree of the gray anomaly index, improve the accuracy of the expression result of the feature vector, and facilitate further searching for the scratch area. Description of the Drawings

[0040] Figure 1 This is the flowchart of a method for detecting the machining of parts on a milling machine according to the present invention. Specific embodiments

[0041] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative efforts fall within the protection scope of the present invention.

[0042] The embodiments of the present invention disclose a method for detecting the machining of parts on a milling machine. Referring to Figure 1 , the following steps are included:

[0043] S1: Obtain a color image of the part machined on the milling machine, and convert the color image to a grayscale image.

[0044] Use an industrial camera to collect the color image of the part after machining on the milling machine. The illumination condition is bright-field illumination, and the image format is the hsv color space model. Obtain the brightness value of each pixel point in the color image, convert the color image to a grayscale image, and obtain the grayscale value of each pixel point in the grayscale image. It should be noted that when there are slight scratches on the surface of the part after machining on the milling machine, the grayscale value of the pixel points in the scratch area is lower than that of the pixel points in the normal area, and in the color image, the brightness value of the pixel points in the scratch area is lower than that of the pixel points in the normal area.

[0045] S2: Calculate the grayscale anomaly index of the pixel points in the grayscale image and the brightness anomaly index of the pixel points in the color image.

[0046] The grayscale anomaly index represents the difference between the grayscale value of the corresponding pixel point and the grayscale values of other pixel points, and the brightness anomaly index represents the difference between the brightness value of the corresponding pixel point and the brightness values of other pixel points.

[0047] Use the canny algorithm to perform edge detection on the grayscale image to obtain edge pixel points, and then remove the edge pixel points to calculate the grayscale anomaly index of the pixel points in the remaining area. The edge pixel points obtained by edge detection are pixel points with significant edge features, that is, the grayscale value of the edge pixel points is significantly lower than that of the pixel points in the normal area, and the grayscale value of the pixel points in the slight scratch area is between the two. Therefore, it is necessary to remove the edge pixel points to reduce the interference of the edge pixel points.

[0048] In one embodiment, the expression of the grayscale anomaly index is:

[0049]

[0050] In the formula, Represents the gray - scale anomaly index of the \(m\) - th pixel point in the gray - scale image, Represents the gray - scale value of the \(m\) - th pixel point in the gray - scale image, Represents the average gray - scale value of the remaining pixel points after removing the edge pixel points in the gray - scale image, Represents the maximum gray - scale value of the neighborhood pixel points of the \(m\) - th pixel point in the gray - scale image. The neighborhood pixel points are eight - neighborhood pixel points. \(a\) represents the number of pixel points with gray - scale values greater than the gray - scale value of the \(m\) - th pixel point among the neighborhood pixel points of the \(m\) - th pixel point, and norm represents the normalization function. This expression is applicable to parts with complex surface contours, such as parts with holes on the surface. Exemplarily, the gray - scale value of the \(m\) - th pixel point is 100, and the gray - scale values of the eight - neighborhood pixel points of the \(m\) - th pixel point are 80, 86, 90, 95, 98, 105, 102, 106 respectively. Among them, the gray - scale values 105, 102, 106 are greater than 100, so the value of \(a\) is 3.

[0051] Represents the difference between the gray - scale value of the corresponding pixel point and the overall average gray - scale value of the gray - scale image. The larger its value, the greater the possibility that the corresponding pixel point is abnormal; Represents the difference between the gray - scale value of the corresponding pixel point and the maximum gray - scale value of the neighborhood pixel points. The larger its value, the greater the difference between the corresponding pixel point and the neighborhood pixel points, and the greater the possibility that the corresponding pixel point is abnormal. The scratches on the surface of the part after milling usually exist in an approximately straight - line manner, and the width of the scratches is small. Therefore, by comparing the gray - scale value of the corresponding pixel point with the gray - scale values of the neighborhood pixel points, it is possible to initially understand whether the corresponding pixel point is a pixel point in the scratch area.

[0052] In one embodiment, the calculation method of the gray - scale anomaly index of the \(m\) - th pixel point is: calculate the absolute value of the difference between the gray - scale value of the \(m\) - th pixel point and the average gray - scale value in the gray - scale image, normalize the absolute value of the difference, and use the normalized result as the gray - scale anomaly index of the \(m\) - th pixel point. This method is applicable to scenarios where the surface structure of the part is simple, such as when the entire machining surface is a plane.

[0053] In summary, the gray - scale anomaly index represents the degree of abnormality of the corresponding pixel point. The larger its value, the greater the possibility that the corresponding pixel point is abnormal, that is, the greater the possibility that the corresponding pixel - point area is the scratch area.

[0054] Similarly, the brightness anomaly index of each pixel point in the color image is obtained. The calculation method of the brightness anomaly index is the same as that of the gray - scale anomaly index, and the specific steps are not elaborated here.

[0055] S3: Construct the feature vector of the corresponding pixel point by using the gray - scale anomaly index and the brightness anomaly index.

[0056] S31: Calculate the weight factor of the gray anomaly index of the pixel point.

[0057] The calculation method is as follows: respectively construct a first window and a second window with the corresponding pixel point as the center, and the area of the second window is larger than that of the first window; construct the first gray-level co-occurrence matrix of the first window, and calculate the first contrast of the first gray-level co-occurrence matrix by using the contrast calculation formula; construct the second gray-level co-occurrence matrix of the second window, and calculate the second contrast of the second gray-level co-occurrence matrix by using the contrast calculation formula; take the ratio of the first contrast to the sum of the first contrast and the second contrast as the weight factor of the gray anomaly index of the corresponding pixel point.

[0058] Exemplarily, construct a 5×5 first window and a 9×9 second window with the m-th pixel point as the center, calculate the first contrast CON1 of the first gray-level co-occurrence matrix, calculate the second contrast CON2 of the second gray-level co-occurrence matrix, and the expression of the weight factor β of the gray anomaly index of the corresponding pixel point is 。

[0059] The contrast of the gray-level co-occurrence matrix reflects the clarity of the image and the depth of the texture grooves. The deeper the texture grooves, the greater the contrast and the clearer the visual effect; conversely, the smaller the contrast, the shallower the grooves and the more blurred the effect. The contrast calculation method is a prior art, and the specific steps and expressions are not elaborated here.

[0060] S32: Construct the feature vector of the corresponding pixel point.

[0061] The expression of the feature vector is:

[0062]

[0063] In the formula, represents the feature vector of the m-th pixel point, represents the gray anomaly index of the m-th pixel point in the gray image, represents the brightness anomaly index of the m-th pixel point in the gray image, and β represents the preset weight factor of the gray anomaly index of the m-th pixel point. Adjusting the gray anomaly index and the brightness anomaly index by using the weight factor improves the accuracy of the feature vector.

[0064] S4: Calculate the scratch index of the pixel point.

[0065] The expression of the scratch index is:

[0066]

[0067] In the formula, represents the scratch index of the m-th pixel point in the gray image, represents the modulus of the feature vector of the m-th pixel point, It represents the modulus of the feature vector of the nth neighboring pixel of the mth pixel, and k represents the total number of neighboring pixels of the mth pixel.

[0068] It reflects the abnormality possibility of the corresponding pixel. It reflects the overall difference between the abnormality possibility of the corresponding pixel and that of the remaining pixels. The scratch index comprehensively reflects the possibility that the corresponding pixel is a pixel in the scratch area. The higher its value, the greater the possibility that the corresponding pixel is a pixel in the scratch area. On the contrary, the lower its value indicates that the corresponding pixel is in the normal area.

[0069] S5: Calculate the scratch threshold, and take the pixels with the scratch index greater than the preset scratch threshold as the marked points.

[0070] Set the scratch threshold as H to obtain multiple marked points. Select any one of the marked points as the center point, and connect the center point with the marked points to obtain multiple connecting lines; Select the line in any one of the directions of 0°, 45°, and 90° as the reference line. In practical applications, according to the actual situation, select the corresponding reference line. Exemplarily, when the scratches are distributed horizontally, select the line in the 0° direction as the reference line; when the scratches are distributed vertically, select the line in the 90° direction as the reference line; when the scratches are distributed obliquely, select the line in the 45° direction as the reference line.

[0071] Construct the objective function, and the expression is: ;

[0072] In the formula, represents the objective function with respect to the scratch threshold H. represents the variance of the cosine values of the angles between multiple connecting lines and the preset reference line. represents the information entropy of the gray value of the pixel corresponding to the marked point, and b represents the number of connecting lines.

[0073] On the surface of the parts processed by the milling machine, the scratches usually exist in an approximate straight line or a straight line manner. represents the overall chaos degree of the connecting lines. The smaller its value, the more regular the arrangement of the connecting lines. represents the contribution degree of each connecting line to the chaos degree. The smaller its value, the more regular the distribution of all connecting lines, and the more in line with the law of the connecting lines showing an approximate straight line structure.

[0074] Use the swarm intelligence optimization algorithm to solve the minimum value of the objective function to obtain the value of H; Repeatedly select multiple marked points as the center points to further obtain multiple values of H, and take the average value of the obtained values of H as the scratch threshold.

[0075] S6: Cluster the marked points to obtain multiple clusters. In response to the number of data points within a cluster being greater than a preset threshold number, issue a prompt indicating that there is a scratch on the part surface.

[0076] Among the obtained multiple marked points, most are pixel points in the scratch area and a part are noise points. Therefore, use the DBSCAN clustering algorithm to cluster the marked points to obtain multiple clusters. Among them, the area corresponding to the cluster with the number of data points exceeding the threshold number is the scratch area. Conversely, the data points within the cluster with a smaller number of data points are noise data points. The threshold number is set manually according to the actual situation. Exemplarily, the threshold number is 8. Use the least squares method to fit the data points within the largest cluster to obtain a fitting line, and further obtain the scratch area, so as to facilitate subsequent processes to process the part surface.

[0077] The embodiment of the present invention also discloses a part milling machine processing detection system, including a processor and a memory. The memory stores computer program instructions, and when the computer program instructions are executed by the processor, a part milling machine processing detection method according to the present invention is implemented.

[0078] The above system also includes other components well-known to those skilled in the art such as a communication bus and a communication interface. Their settings and functions are known in the art, so they will not be elaborated here.

[0079] The above are all preferred embodiments of the present invention. The protection scope of the present invention is not limited by this. Therefore, all equivalent changes made according to the structure, shape, and principle of the present invention should be covered within the protection scope of the present invention.

Claims

1. A method for machining and detecting parts by a milling machine, characterized in that, Including the steps: Obtain a color image of a part processed by a milling machine, and convert the color image to obtain a grayscale image; Calculate the grayscale anomaly index of the pixel points in the grayscale image. The grayscale anomaly index is positively correlated with the absolute value of the difference between the grayscale value of the corresponding pixel point and the grayscale values of other pixel points. Calculate the brightness anomaly index of the pixel points in the color image. The brightness anomaly index is positively correlated with the absolute value of the difference between the brightness value of the corresponding pixel point and the brightness values of other pixel points. Construct a feature vector for the corresponding pixel point using the grayscale anomaly index and the brightness anomaly index. Calculate the scratch index of the pixel point, and use the pixel points with a scratch index greater than a preset scratch threshold as marked points. Cluster the marked points to obtain multiple clusters. In response to the number of data points in the cluster being greater than a preset number threshold, issue a prompt that there are scratches on the part surface; Among them, the expression of the scratch index is: In the formula, represents the scratch index of the m-th pixel point in the grayscale image, represents the modulus of the feature vector of the m-th pixel point, represents the modulus of the feature vector of the n-th neighboring pixel point of the m-th pixel point, and k represents the total number of neighboring pixel points of the m-th pixel point; The calculation method of the grayscale anomaly index is: perform edge detection on the grayscale image to obtain edge pixel points. The expression of the grayscale anomaly index is: In the formula, represents the gray anomaly index of the m-th pixel point in the grayscale image, represents the gray value of the m-th pixel point in the grayscale image, represents the average gray value of the remaining pixel points after removing the edge pixel points in the grayscale image, represents the maximum gray value of the neighborhood pixel points of the m-th pixel point in the grayscale image, a represents the number of neighborhood pixel points of the m-th pixel point whose gray value is greater than the gray value of the m-th pixel point, and norm represents the normalization function; The method further includes: respectively construct a first window and a second window centered on the corresponding pixel point. The area of the second window is larger than the area of the first window; Construct a first gray-level co-occurrence matrix of the first window, and calculate the first contrast of the first gray-level co-occurrence matrix using the contrast calculation formula. Construct a second gray-level co-occurrence matrix of the second window, and calculate the second contrast of the second gray-level co-occurrence matrix using the contrast calculation formula. Use the ratio of the first contrast to the sum of the first contrast and the second contrast as the weight factor of the grayscale anomaly index of the corresponding pixel point.

2. The machining detection method of a part milling machine according to claim 1, characterized in that The expression of the feature vector is: In the formula, represents the feature vector of the m-th pixel point, represents the gray anomaly index of the m-th pixel point in the grayscale image, represents the lightness anomaly index of the m-th pixel point in the grayscale image, and β represents the weight factor of the preset gray anomaly index of the m-th pixel point.

3. A method for machining and detecting a part milling machine according to claim 1, characterized in that, The calculation method of the scratch threshold is: set the scratch threshold to H, obtain multiple marked points, select any one of the marked points as the center point, and connect the center point with the marked points to obtain multiple connecting lines; Construct the objective function, and the expression is: ; In the formula, represents the objective function with respect to the scratch threshold H, represents the variance of the cosine values of the angles between multiple connecting lines and a preset reference line, represents the information entropy of the gray values of the pixel points corresponding to the marked points, and b represents the number of connecting lines; solve the minimum value of the objective function to obtain the value of H; repeatedly select multiple marked points as the center points to further obtain multiple values of H, and take the average value of the obtained values of H as the scratch threshold.

4. A method for machining and detecting parts on a milling machine according to claim 3, characterized in that, The selection method of the reference line is: select a line in any one of the directions of 0°, 45°, and 90° as the reference line.

5. A method for machining and detecting parts on a milling machine according to claim 1, characterized in that Use a swarm intelligence optimization algorithm to solve the minimum value of the objective function.

6. A method for machining and detecting a part milling machine according to claim 1, characterized in that The method further includes: fitting the data points in the largest cluster using the least squares method to obtain a fitting line, and further obtaining the scratch area.

7. A method for machining and detecting a part milling machine according to claim 1, characterized in that, The method further includes: performing edge detection on the grayscale image using the canny algorithm to obtain edge pixel points.

8. A part milling machine processing and inspection system, characterized in that, Including: A processor and a memory. The memory stores computer program instructions. When the computer program instructions are executed by the processor, a method for detecting the machining of parts by a milling machine according to any one of claims 1-7 is implemented.

Images