Cutting process monitoring system

Abstract

To provide a cutting work monitoring system which can highly accurately analyze the state of at least one of a tool blade tip during cutting and a processing surface of a cutting object material in real time.SOLUTION: A cutting work monitoring system comprises: a camera 23 which is positioned with respect to a lathe tool 21 such that a tool blade tip in an imaging visual field becomes the same position and consecutively images the tool blade tip; a classification unit 33 which classifies the plurality of images captured by the camera 23 into any of a plurality of patterns on the basis of information reflected in each image; and an analysis unit 34 which analyzes the state of at least one of the tool blade tip and the processing surface of a cutting object material W from the image classified into the prescribed pattern among the plurality of patterns.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] The present invention relates to a cutting process monitoring system used, for example, to prevent troubles in cutting processes.

Background Art

[0002] In lathe machining, it is generally necessary to quickly detect the occurrence of wear or chipping of the tool edge and take corresponding measures such as replacing the tool. For this reason, conventionally, a method of photographing the tool edge with a camera and examining the state of the edge from the image has been known. For example, in Patent Document 1, the tool edge is photographed with a camera in the standby state (stationary state) before and after cutting, and the wear amount is determined from the binary image. Further, in Patent Document 2, the tool edge during cutting is captured with an infrared camera, and an abnormality determination is made by learning it with a neural network.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0004] In the method of Patent Document 1, since the tool edge is photographed with a camera in the standby state (stationary state) other than during cutting, there is a problem that it is not possible to immediately respond to the occurrence of troubles during cutting. In the method of Patent Document 2, since it is an indirect analogy of the tool wear amount from the temperature distribution, it is difficult to quantitatively evaluate the wear amount, and a lot of learning is required to improve the determination accuracy.

[0005] One object of the present invention is to provide a cutting process monitoring system that can accurately analyze the state of at least one of the tool edge and the machined surface of the workpiece during cutting in real time.

Means for Solving the Problem

[0006] One aspect of the cutting process monitoring system of the present invention is positioned with respect to a turning tool so that the tool tip within the imaging field of view is at the same position, and the tool tip during turning and the machined surface of the workpiece to be machined is continuously photographed possible a camera, a classification unit that classifies a plurality of images captured by the camera into any one of a plurality of patterns based on the information shown in each of the images, and from the images classified into a predetermined pattern among the plurality of patterns, an analysis unit that analyzes at least one of the states of the tool tip and the machined surface of the workpiece, and The plurality of images include an image in which the machined surface is shown and an image in which the machined surface is not shown. the plurality of patterns include a pattern in which an image showing the machined surface of the workpiece is classified.

[0007] In the present invention, the tool tip and the machined surface of the workpiece during turning (hereinafter, may be simply abbreviated as the tool tip, etc.) are continuously photographed by a camera, that is, in real time. Note that "continuously photographing" as used in the present invention means photographing at least 10 frames or more per second. In addition, in order to position and fix the camera with respect to the turning tool so that the tool tip within the imaging field of view is at the same position, for example, the camera may be attached to the tool post that supports the turning tool via an arm or the like. Then, the classification unit classifies each image into any one of a plurality of patterns by comparing the obtained image information with, for example, the learning result of the storage unit. The analysis unit analyzes at least one of the states of the tool tip and the machined surface of the workpiece from each classified image.

[0008] Specifically, among the images continuously captured by the camera, there are those in which the tool cutting edge is hidden by chips, those with a lot of welding adhering to the cutting edge and the wear state cannot be seen, and so on. Therefore, from among a large number of consecutively taken images, for example, those in which the contour of the cutting edge (the cutting edge ridge line), welding to the rake face, etc., or the machined surface of the workpiece can be seen are selected (classified), and the state of wear and defects of the cutting edge, the state of welding, and the machining state of the machined surface are analyzed by image processing. Deep learning can be used for the above selection, and for relatively simple determination, a high determination accuracy can be obtained with a small amount of learning.

[0009] As described above, according to the present invention, at least one of the state of the tool cutting edge and the machined surface of the workpiece during cutting can be analyzed with high accuracy in real time. That is, a user or the like can grasp the state of the tool cutting edge or the like during cutting in real time. Thereby, for example, the practicality as an abnormality detection system or an in-process (during cutting) machining dimensional accuracy measurement system is enhanced.

[0010] In the above cutting process monitoring system, the classification unit classifies the image in which the cutting edge ridge line of a predetermined cutting edge length or more appears into the first pattern among the plurality of patterns, and the analysis unit preferably analyzes the damage state of the tool cutting edge based on the shape of the cutting edge ridge line of the image classified into the first pattern.

[0011] In this case, by the analysis unit analyzing the image classified into the first pattern by the classification unit, the damage state and damage amount such as wear and defects of the tool cutting edge can be measured. Specifically, for example, the cutting edge ridge line (edge) is detected by image processing such as a canny filter, and the change (increase amount) in the number of pixels of the edge is measured to analyze the damage state of the cutting edge.

[0012] In the above cutting process monitoring system, the classification unit classifies the image in which welding appears on the tool cutting edge into the second pattern among the plurality of patterns, and the analysis unit preferably analyzes the damage state of the tool cutting edge based on the change in the appearance frequency of the image classified into the second pattern.

[0013] Generally, when damage occurs at the tool tip, welding tends to occur due to this damage. That is, during cutting, welding to and peeling (detachment) from the tool tip are repeated randomly. However, when damage occurs at the tool tip, welding often adheres strongly to the damaged part and is difficult to remove, resulting in a high frequency of appearance of welding in the image. That is, for example, when the continuous shooting speed (shooting interval) of the camera is constant, the number of images classified into the second pattern per unit time increases. According to the above configuration of the present invention, even when the tool tip is hidden or difficult to see due to welding, damage to the tool tip can be indirectly detected. That is, based on the change in the appearance frequency of the images classified into the second pattern by the classification unit, the analysis unit can estimate the damage state of the tool tip.

[0014] In the above cutting process monitoring system, it is preferable that the classification unit classifies the image in which the machined surface of the workpiece is shown into a third pattern among the plurality of patterns, and the analysis unit analyzes the machining state of the machined surface from the image classified into the third pattern.

[0015] In this case, by analyzing the images classified into the third pattern by the classification unit by the analysis unit, it is possible to determine the machining dimensional accuracy, surface finish quality, burr generation state, and chip discharge state of the machined surface of the workpiece. That is, the "machining state of the machined surface of the workpiece" referred to in the present invention refers to any one or more of the machining dimensional accuracy, surface finish quality, burr generation state, and chip discharge state of the machined surface of the workpiece. In particular, with the above configuration of the present invention, the user can grasp the machining accuracy and quality of the machined surface immediately after cutting in real time, and for example, can quickly take measures when problems occur in the machining state.

[0016] In the above cutting process monitoring system, it is preferable that the classification unit performs image classification by deep learning.

[0017] In this case, the classification determination accuracy by the classification unit can be stably improved by deep learning.

Advantages of the Invention

[0018] According to the cutting process monitoring system of one aspect of the present invention, at least one of the state of the cutting tool edge and the machined surface of the workpiece during cutting can be analyzed with high precision in real time.

Brief Description of the Drawings

[0019]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Embodiments for Carrying Out the Invention

[0020] The cutting process monitoring system 10 of an embodiment of the present invention will be described with reference to the drawings. The cutting process monitoring system 10 of the present embodiment includes a lathe 20 and a cutting process monitoring device 30.

[0021] As shown in FIGS. 1 and 2, the turning device 20 is a machine tool such as a numerically controlled lathe. The turning device 20 is a device for turning a workpiece W made of metal or the like. That is, the cutting process monitoring system 10 is used for turning (lathe work) by a machine tool such as a numerically controlled lathe. Turning refers to cutting with a turning tool such as a tool bit.

[0022] The turning device 20 includes a turning tool 21, a tool rest 22, a camera 23, an arm 24, and a chuck (not shown). That is, the cutting process monitoring system 10 includes a camera 23.

[0023] As shown in FIG. 1, the turning tool 21 is, for example, a tool bit with an exchangeable cutting edge in which a cutting insert 26 is detachably attached to the tip of a holder 25. That is, the turning tool 21 has a holder 25 and a cutting insert 26. Note that the turning tool 21 is not limited to this configuration and may be, for example, a solid type tool bit in which the tool cutting edge is integrally formed with the holder.

[0024] The holder 25 is columnar and extends in one direction. In the example shown in FIG. 1, the one direction in which the holder 25 extends is inclined at an angle θ with respect to the horizontal plane H. The angle θ is, for example, 30°. The rear end portion of the holder 25 is supported by the tool rest 22. The turning tool 21 and the tool rest 22 are integrally fixed.

[0025] The cutting insert 26 is made of, for example, cemented carbide. The cutting insert 26 is, for example, a polygonal plate shape such as a square plate shape, or a disc shape. In this embodiment, the cutting insert 26 is, for example, a rhombic plate shape. As shown in FIG. 3, the cutting insert 26 has a rake face 26a, a flank face (not shown), and a cutting edge 26b disposed at the ridge line portion between the rake face 26a and the flank face. In this embodiment, at least a part of the cutting insert 26, specifically, the part including at least the rake face 26a and the cutting edge 26b, may be simply referred to as the "tool cutting edge" or the "cutting edge".

[0026] The rake face 26a is disposed on at least one of the pair of plate surfaces facing the plate thickness direction of the cutting insert 26. As shown in FIG. 1, the rake face 26a is disposed on one of the plate surfaces facing the upper side of the cutting insert 26, that is, the upper surface. In the example shown in FIG. 1, the rake face 26a is inclined with respect to the horizontal plane H at substantially the same angle as the angle θ at which the holder 25 is inclined with respect to the horizontal plane H.

[0027] As shown in FIG. 5, the rake face 26a of the present embodiment has a feature point 26c. The feature point 26c constitutes a part of the pattern provided on the rake face 26a. The feature point 26c is provided on the rake face 26a, for example, for the purpose of smoothly discharging the chip C, enhancing the appearance design, and enhancing the discriminability. In the present embodiment, the feature point 26c is, for example, a protruding chip breaker or the like.

[0028] As shown in FIG. 1, the cutting edge 26b is disposed at the tip of the turning tool 21. By bringing the cutting edge 26b into contact with the workpiece W rotating around the central axis O, turning processing is performed on the workpiece W, and a processed surface Wa as shown in FIG. 5 is formed.

[0029] In the present embodiment, as shown in FIG. 3, when looking at the upper surface of the cutting insert 26, that is, the rake face 26a, from the front, the cutting edge 26b is substantially V-shaped. Specifically, the cutting edge 26b has one corner edge extending in a curved shape and a pair of straight edges connected to both ends of the corner edge and extending linearly.

[0030] As shown in FIG. 1, the tool post 22 supports the turning tool 21. The tool post 22 moves the turning tool 21 with respect to the workpiece W at least in the direction in which the horizontal plane expands, that is, along the plane direction of the horizontal plane. The tool post 22 may move the turning tool 21 with respect to the workpiece W in the vertical direction.

[0031] The camera 23 is disposed above the turning tool 21 and the workpiece W. The camera 23 is, for example, a camera of a global shutter system. The camera 23 continuously photographs the tool tip. Note that "continuously photographing" as used in this embodiment refers to photographing at least 10 frames or more per second. Specifically, in this embodiment, the camera 23 takes successive shots of the tool tip at a speed of, for example, 30 frames or more per second. During cutting, the camera 23 also photographs the machined surface Wa of the workpiece W together with the tool tip.

[0032] In this embodiment, the camera 23 photographs the tool tip from a direction perpendicular to the rake face 26a. Specifically, the camera 23 photographs the tool tip from a direction inclined by a predetermined angle with respect to the vertical direction. This predetermined angle is substantially the same as the angle at which the rake face 26a is inclined with respect to the horizontal plane H (corresponding to the angle θ). The distance between the camera 23 and the tool tip is, for example, 300 mm or more.

[0033] The camera 23 is fixed to the tool rest 22 via the arm 24. Therefore, the camera 23 can move following the movement of the tool rest 22. The camera 23 is positioned with respect to the turning tool 21 so that the tool tip within the imaging field is at the same position, as shown in FIGS. 3 to 5, for example.

[0034] As shown in FIG. 1, the arm 24 connects the camera 23 and the tool rest 22. The front end portion of the arm 24 is connected to the camera 23, and the rear end portion is connected to the tool rest 22. The arm 24 includes at least two or more shaft portions 24a, at least one or more joint portions 24b that connect the ends of adjacent shaft portions 24a, and a fixed base 24c that fixes the rear end portion of the arm 24 to the tool rest 22.

[0035] The shaft portion 24a is, for example, a shaft or a pipe. The joint part 24b can be switched between a lock mode in which the ends of the shaft part 24a are non-rotatably fixed by operating a knob or the like (not shown), and a free mode in which the ends of the shaft part 24a are rotatably connected. By providing the joint part 24b, the arm 24 can be deformed. By changing the shape of the arm 24, the camera 23 can be position-adjusted to face the tool cutting edge regardless of, for example, the tool shape and type of the turning tool 21. During cutting, the joint part 24b is set in the lock mode.

[0036] The fixing base 24c fixes the arm 24 to the tool rest 22 by, for example, magnetic force or the like. Therefore, the arm 24 can adjust its attachment position with respect to the tool rest 22. That is, in the present embodiment, by adjusting the attachment position of the fixing base 24c to the tool rest 22, the camera 23 can also be position-adjusted to face the tool cutting edge.

[0037] Although not particularly shown, the chuck detachably holds the workpiece W. The chuck rotates the workpiece W around its central axis O.

[0038] Although not particularly shown, the turning device 20 may be provided with a backlight for emphasizing the contour of the cutting edge 26b and the machined surface Wa, and a front light for illuminating the rake face 26a. Thereby, the shutter speed of the camera 23 may be increased further.

[0039] As shown in FIG. 2, the cutting process monitoring device 30 includes an image acquisition unit 31, a storage unit 32, a classification unit 33, an analysis unit 34, and a display unit 35. That is, the cutting process monitoring system 10 includes the classification unit 33 and the analysis unit 34.

[0040] The image acquisition unit 31 acquires an image captured by the camera 23, that is, image data. The image acquisition unit 31 stores the acquired image information in the storage unit 32. Note that the cutting process monitoring device 30 may store, for example, the acquired image information in association with a tool ID, a device ID, a user ID, etc. in the storage unit 32.

[0041] The storage unit 32 stores various information used by the cutting process monitoring device 30. The storage unit 32 stores the acquired image information. The storage unit 32 stores, for example, deep learning teacher data, learning results, etc. used by the classification unit 33 to classify images.

[0042] The classification unit 33 executes image classification by deep learning, for example, based on the learning results stored in the storage unit 32. That is, the classification unit 33 classifies a plurality of images captured by the camera 23 into any one of a plurality of patterns based on the information shown in each image. In the present embodiment, the plurality of patterns include a first pattern (cutting edge ridge line), a second pattern (welding), and a third pattern (machined surface of the workpiece).

[0043] Specifically, as shown in FIG. 3, the classification unit 33 classifies an image in which a cutting edge ridge line of a predetermined blade length or more of the cutting edge 26b is shown in the imaging field of view into the first pattern among the plurality of patterns. In an example shown in FIG. 3, the cutting edge 26b is not in contact with the workpiece W, and the entire cutting edge ridge line is clearly shown in the imaging field of view.

[0044] Also, as shown in FIG. 4, the classification unit 33 classifies an image in which welding TA is shown on the tool cutting edge into the second pattern among the plurality of patterns. In an example shown in FIG. 4, at least a part of the rake face 26a and at least a part of the cutting edge 26b are hidden by the welding TA attached to the tool cutting edge.

[0045] Also, as shown in FIG. 5, the classification unit 33 classifies an image in which the machined surface Wa of the workpiece W is shown into the third pattern among the plurality of patterns. In an example shown in FIG. 5, the surface and contour of the machined surface Wa immediately after being cut by the cutting edge 26b are clearly shown. Each image shown in FIGS. 3 to 5 is an image of the tool cutting edge taken when turning the plate-shaped workpiece W as shown in FIG. 1. At this time, the material of the workpiece W is stainless steel, the rotational speed of the workpiece W is 100 revolutions per minute, and the cutting was performed in a dry environment.

[0046] The classification of the images by the classification unit 33 is performed, for example, by the following method. First, in the test cutting performed before the main cutting of the workpiece W, a plurality of images are taken by the camera 23. The plurality of images thus obtained are classified and stored into a first pattern (cutting edge ridge line) in which the cutting edge 26b is not in contact with the workpiece W and the cutting edge ridge line is clearly visible, a second pattern (welding) in which the cutting edge 26b is not in contact with the workpiece W but the welding TA is placed on the rake face 26a, and a third pattern (machined surface of the workpiece) that captures the moment when the cutting edge 26b comes into contact with the workpiece W (teaching data), and this classification method is learned by deep learning. That is, based on the teaching data in the storage unit 32, machine learning for estimating the classification of the first to third patterns is executed, and the determination accuracy of the image classification is improved by, for example, feeding back the learning result to the storage unit 32. Note that this classification is an example, and what kind of classification method is appropriate varies depending on the cutting conditions, the type and shape of the workpiece W, and the like. In addition, for the classification (sorting) of the above-described image data, image automatic recognition software such as HALCON (manufactured by LINKS Co., Ltd.) can be used.

[0047] The analysis unit 34 analyzes at least one of the state of the tool cutting edge and the machined surface Wa of the workpiece W from the images classified into a predetermined pattern among the plurality of patterns. As the analysis unit 34, for example, an image processing program or the like can be used.

[0048] Specifically, the analysis unit 34 analyzes the damage state of the tool cutting edge based on the shape of the cutting edge ridge line of the cutting edge 26b in the image classified into the first pattern (hereinafter sometimes referred to as the first analysis). In the present embodiment, the analysis unit 34 detects the cutting edge ridge line (edge) of the cutting edge 26b by image processing such as a canny filter, for example, and analyzes the damage state of the cutting edge by measuring the change (increase amount) in the number of pixels of the edge.

[0049] Figures 6 and 7 show an example of image processing of the cutting edge ridge line of the cutting edge 26b by the analysis unit 34. In a predetermined cutting process, the image in FIG. 7 (original image) was taken later than the image in FIG. 6. In FIG. 7 compared with FIG. 6, convex portions due to welding TA and concave portions due to chipping CP are formed on the cutting edge ridge line. Due to the unevenness, the number of pixels of the edge, that is, the contour length of the cutting edge ridge line, has become longer. That is, by comparing the number of pixels of the edge, the damage state of the cutting edge can be estimated. In this way, the analysis unit 34 analyzes the damage state of the tool cutting edge. Note that the above analysis is an example, and other analysis methods may be used for the analysis of the first pattern.

[0050] Further, the analysis unit 34 analyzes the damage state of the tool cutting edge based on the change in the appearance frequency of the images classified into the second pattern (hereinafter sometimes referred to as the second analysis). FIG. 8 shows an example of the change in the appearance frequency of the images classified into the second pattern. The hatched portion in the graph of FIG. 8 indicates that the image has been classified into the second pattern (welding) by the classification unit 33.

[0051] Specifically, during cutting, welding to the tool cutting edge and its peeling (falling off) are repeated randomly. However, when damage occurs to the cutting edge, welding often adheres strongly to the damaged portion and is difficult to remove, so the occurrence frequency of welding increases. For this reason, in FIG. 8, it is presumed that cutting edge damage has occurred in the vicinity of the symbol P on the horizontal axis (cutting time). In this way, the analysis unit 34 analyzes the damage state of the tool cutting edge. Note that the above analysis is an example, and other analysis methods may be used for the analysis of the second pattern.

[0052] Further, the analysis unit 34 analyzes the machining state of the machined surface Wa of the workpiece W from the images classified into the third pattern (hereinafter sometimes referred to as the third analysis). Note that the "machining state of the machined surface Wa of the workpiece W" as used in the present embodiment refers to any one or more of the machining dimensional accuracy, surface finish, burr generation state, and chip C discharge state of the machined surface Wa of the workpiece W.

[0053] In FIG. 5, the analysis unit 34 determines the machining dimensional accuracy by measuring, through image processing, the distance L between, for example, the feature point 26c on the scooping surface 26a and the machined surface Wa. That is, in the present embodiment, even if blurring (displacement of the origin position) occurs in the camera 23 due to, for example, vibration during cutting, the machining dimensions can be measured with high accuracy. Further, the analysis unit 34 determines the surface quality of the machined surface by measuring, for example, the dimensions of the uneven shape appearing on the surface or contour of the machined surface Wa. Further, the analysis unit 34 may analyze, for example, the presence or absence of burr generation on the machined surface Wa, the length and curl state of the chip C, and the discharge direction of the chip C. In this way, the analysis unit 34 analyzes the machining state of the machined surface Wa of the workpiece W. Note that the above analysis is an example, and other analysis methods may be used for the analysis of the third pattern.

[0054] The display unit 35 displays various information of the cutting machining monitoring system 10. For example, the analysis results of the above-described analysis unit 34 are displayed on the display unit 35.

[0055] FIG. 9 is a flowchart showing an example of the processing of the cutting machining monitoring device 30. As shown in FIG. 9, first, the image acquisition unit 31 acquires the image information output from the camera 23 (step S1). Next, the classification unit 33 classifies each image into the first to third patterns based on the learning result stored in the storage unit 32 (step S2). That is, the classification unit 33 classifies each image into any one of the first to third patterns by comparing the acquired image information with the learning result.

[0056] Next, the analysis unit 34 analyzes the state of at least one of the tool cutting edge and the machined surface Wa of the workpiece W from the images classified into the first to third patterns (step S3). Specifically, the analysis unit 34 performs the first analysis as described above on the image classified into the first pattern (cutting edge ridge line). The analysis unit 34 also performs the second analysis as described above on the image classified into the second pattern (welding). The analysis unit 34 also performs the third analysis as described above on the image classified into the third pattern (machined surface of the workpiece). Next, the display unit 35 displays the analysis result output from the analysis unit 34 (step S4). After the process of step S4, the cutting process monitoring device 30 ends the process of the above example.

[0057] In the cutting process monitoring system 10 of the present embodiment described above, the tool cutting edge and the machined surface Wa of the workpiece W (hereinafter sometimes simply referred to as the tool cutting edge, etc.) during turning are continuously photographed by the camera 23, that is, in real time. The camera 23 is positioned and fixed with respect to the turning tool 21 so that the tool cutting edges within the imaging field of view are in the same position. Then, the classification unit 33 classifies each image into one of a plurality of patterns by comparing the obtained image information with the learning result of the storage unit 32. The analysis unit 34 analyzes the state of at least one of the tool cutting edge and the machined surface Wa of the workpiece W from each classified image.

[0058] Specifically, among the images continuously photographed by the camera 23, there are those in which the tool cutting edge is hidden by the chip C, those in which a large amount of welding TA adheres to the cutting edge and the wear state cannot be seen, and the like. Therefore, in the present embodiment, from among a large number of continuously taken images, those in which the contour of the cutting edge 26b (cutting edge ridge line), the welding TA on the rake face 26a, or the machined surface Wa of the workpiece W can be seen are selected (classified), and the wear and defect states of the cutting edge 26b are analyzed by image processing, the state of the welding TA is analyzed, or the machining state of the machined surface Wa is analyzed. Deep learning can be used for the above selection, and for relatively simple determination, a high determination accuracy can be obtained with a small amount of learning.

[0059] According to the present embodiment as described above, it is possible to analyze the state of at least one of the tool edge tip during cutting and the machined surface Wa of the workpiece W with high precision in real time. That is, a user or the like can grasp the state of the tool edge tip during cutting in real time. As a result, for example, the practicality as an abnormality detection system or an in-process (during cutting) machining dimension accuracy measurement system is enhanced.

[0060] In the present embodiment, the classification unit 33 classifies an image in which an edge ridge line of a tool edge with a predetermined edge length or more is captured within the imaging field of view into a first pattern among a plurality of patterns, and the analysis unit 34 analyzes the damage state of the tool edge based on the shape of the edge ridge line of the image classified into the first pattern. In this case, by the analysis unit 34 analyzing the image classified into the first pattern by the classification unit 33, it is possible to measure the damage state and damage amount such as wear and chipping of the tool edge tip.

[0061] In the present embodiment, the classification unit 33 classifies an image in which welding TA is captured on the tool edge tip into a second pattern among a plurality of patterns, and the analysis unit 34 analyzes the damage state of the tool edge based on the change in the appearance frequency of the image classified into the second pattern. Generally, when damage occurs to the tool edge tip, there is a tendency for welding TA to be likely to occur due to this damage. That is, when damage occurs to the edge tip, the welding TA strongly adheres to the damaged portion and is difficult to remove, so the appearance frequency of the welding TA in the image increases. That is, for example, when the continuous shooting speed (shooting interval) of the camera 23 is constant, the number of images classified into the second pattern per unit time increases. According to the present embodiment, even when the tool edge tip is hidden by welding TA and is difficult to see, the damage to the edge tip can be indirectly detected. That is, based on the change in the appearance frequency of the image classified into the second pattern by the classification unit 33, the analysis unit 34 can estimate the damage state of the tool edge tip.

[0062] Also, in the present embodiment, the classification unit 33 classifies an image in which the machined surface Wa of the workpiece W is shown into a third pattern among a plurality of patterns, and the analysis unit 34 analyzes the machining state of the machined surface Wa from the image classified into the third pattern. In this case, by the analysis unit 34 analyzing the image classified into the third pattern by the classification unit 33, it is possible to determine the machining dimensional accuracy, the surface quality of the machined surface, the occurrence state of burrs, and the discharge state of the chips C, etc. of the machined surface Wa of the workpiece W. In particular, in the present embodiment, the user or the like can grasp in real time the machining accuracy and quality of the machined surface Wa immediately after cutting, and for example, can quickly take measures when problems occur in the machining state.

[0063] Also, in the present embodiment, the classification unit 33 performs image classification by deep learning. In this case, the determination accuracy of the classification by the classification unit 33 is stably enhanced by deep learning.

[0064] Also, in the present embodiment, the camera 23 photographs the tool cutting edge from a direction perpendicular to the rake face 26a, that is, from a direction directly facing the rake face 26a. In this case, the camera 23 can easily focus on the tool cutting edge, and the effects of the present embodiment described above can be obtained more stably.

[0065] Also, in the present embodiment, the distance between the camera 23 and the tool cutting edge is 300 mm or more. When the above distance is 300 mm or more, for example, when replacing the cutting insert 26 with a new one or indexing the cutting edge 26b (changing the used corner), that is, when changing the cutting edge, the camera 23 is less likely to interfere with the work. Also, even when the chips C extend or scatter during cutting, it is difficult for the chips C to contact the camera 23, so an effective image can be stably acquired.

[0066] Note that the present invention is not limited to the above-described embodiment, and for example, as described below, configuration changes and the like are possible without departing from the gist of the present invention.

[0067] In the foregoing embodiment, an example was given in which the camera 23 is fixed to the tool rest 22 via an arm 24 that can be switched between a lock mode and a free mode, but the present invention is not limited to this. That is, the camera 23 may be fixed to the tool rest 22 via a rigid member or the like that cannot be deformed. Further, the camera 23 may be directly fixed to the tool rest 22. Since the camera 23 only needs to be positioned with respect to the turning tool 21 so that the tool tip within the imaging field of view is at the same position, the camera 23 may be fixed to a member such as a holder 25 (turning tool 21) other than the tool rest 22.

[0068] In addition, within the scope not departing from the gist of the present invention, the respective configurations described in the foregoing embodiments and modification examples may be combined, and addition, omission, substitution, and other changes of the configuration are possible. Further, the present invention is not limited by the foregoing embodiments, but is limited only by the scope of the claims.

Industrial Applicability

[0069] According to the cutting process monitoring system of the present invention, at least one of the states of the tool tip during cutting and the machined surface of the workpiece can be analyzed with high accuracy in real time. Therefore, it has industrial applicability.

Explanation of Reference Numerals

[0070] 10…Cutting process monitoring system, 21…Turning tool, 23…Camera, 33…Classification unit, 34…Analysis unit, TA…Welding, W…Workpiece, Wa…Machined surface

Claims

**Claim 1** A camera positioned with respect to a turning tool such that the tool tip within the imaging field of view is at the same position, and capable of continuously photographing the tool tip and the machined surface of the workpiece during turning; A classification unit that classifies a plurality of images captured by the camera into any one of a plurality of patterns based on the information shown in each of the images; An analysis unit that analyzes at least one of the states of the tool tip and the machined surface of the workpiece from the images classified into a predetermined pattern among the plurality of patterns, comprising: The plurality of images include an image in which the machined surface is shown and an image in which the machined surface is not shown; The plurality of patterns include a pattern in which an image showing the machined surface of the workpiece is classified; A cutting process monitoring system. **Claim 2** The classification unit classifies the image in which the tool tip edge line of a predetermined edge length or more is shown into a first pattern among the plurality of patterns; The analysis unit analyzes the damage state of the tool tip based on the shape of the tool tip edge line of the image classified into the first pattern; The cutting process monitoring system according to Claim 1. **Claim 3** The classification unit classifies the image in which welding is shown on the tool tip into a second pattern among the plurality of patterns; The analysis unit analyzes the damage state of the tool tip based on the change in the appearance frequency of the image classified into the second pattern; The cutting process monitoring system according to Claim 1 or 2. **Claim 4** The classification unit classifies the image in which the machined surface of the workpiece is shown into a third pattern among the plurality of patterns; The analysis unit analyzes the machining state of the machined surface from the image classified into the third pattern; The cutting process monitoring system according to any one of Claims 1 to 3. **Claim 5** The classification unit performs image classification by deep learning; The cutting process monitoring system according to any one of Claims 1 to 4.

Images