Shape extraction method, abrasion detection method, shape extraction device, and abrasion detection device

Abstract

To provide a shape extraction method and the like that can accurately extract the shape of an extraction object part.SOLUTION: A shape extraction method for imaging a tool including a first region and a second region adjacent to the first region, and for extracting the shape of an extraction object part at least part of which is positioned at a boundary part between the first region and the second region includes: an irradiation step of radiating irradiation light toward the tool and a reflection surface; an imaging step of imaging an image including the extraction object part by an imaging part; and an extraction step of extracting the shape of the extraction object part from an image that is imaged in the imaging step. The reflection surface has a color different from the color of the irradiation light and the color of the tool. The irradiation light is reflected toward the second region by the reflection surface. In the extraction step, the shape of the extraction object part is extracted from the image, based on the light strength of a predetermined color in the image.SELECTED DRAWING: Figure 7

Description

[Technical field]

[0001] The present invention relates to a shape extraction method, a wear detection method, a shape extraction device, and a wear detection device. [Background technology]

[0002] Since image measurement can easily detect the boundaries and edges of objects from image information, it is applied to automatic measurement technology for tool wear and tool breakage.

[0003] For example, in the technology disclosed in Patent Document 1, a vertical epi-illumination light source and a half mirror are combined and a target tool is photographed by a CCD camera. Then, based on the binarized image, the graphite oxide part of the tool is extracted.

[0004] In the technique disclosed in Patent Document 2, the tool is observed in an observation system in which a light source irradiating the tool, a camera, and the tool wear surface are arranged in a regular reflection relationship, and the area of ​​the wear surface is detected based on the difference in image brightness. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] JP 2017-83201 A [Patent Document 2] JP 2020-153730 A [Patent Document 3] Patent Publication No. 2021-15616 Summary of the Invention [Problem to be solved by the invention]

[0006] In the techniques disclosed in Patent Documents 1 and 2, the portion to be extracted (extraction target portion) is affected by a slope adjacent to the portion, which causes a problem of poor shape extraction accuracy.

[0007] In response to this, Patent Document 2 proposes an adjustment mechanism for adjusting the angle that the light source and camera make with respect to the extraction target portion, or a method for adjusting by providing multiple light sources and cameras so as to obtain a specular reflection image of the extraction target portion. However, in this case, it becomes necessary to provide these adjustment mechanisms, light sources, and cameras additionally.

[0008] The technology disclosed in Patent Document 3 is also known, but in this technology, in order to efficiently extract boundary or edge information around an object, it is necessary to obtain multiple images by changing the image capture mode, for example by installing two cameras.

[0009] The present invention has been made in consideration of the above circumstances, and aims to provide a shape extraction method and shape extraction device capable of clearly extracting the shape of the extraction target part, as well as a wear detection method and wear detection device capable of accurately detecting wear on the extraction target part. [Means for solving the problem]

[0010] In order to solve the above problem, a shape extraction method according to a first aspect of the present invention is a shape extraction method for photographing a tool including a first area and a second area adjacent to the first area, and extracting the shape of an extraction target part, at least a portion of which is located at the boundary between the first area and the second area, the shape extraction method comprising: a preparation step of preparing a reflection part having a reflective surface; an irradiation step of irradiating light toward the tool and the reflection surface; a photographing step of photographing an image including the extraction target part by a photographing unit; and an extraction step of extracting the shape of the extraction target part from the image photographed in the photographing step, wherein the reflection surface has a color different from both the color of the irradiated light and the color of the tool, and the irradiated light is reflected by the reflection surface toward the second area, and in the extraction step, the shape of the extraction target part is extracted from the image based on the light intensity of a predetermined color in the image.

[0011] A second aspect of the present invention is directed to the shape extraction method of the first aspect, wherein in the preparation step, the reflecting portion is disposed so that the reflecting surface is located behind the tool when viewed from the photographing direction.

[0012] Furthermore, in aspect 3 of the present invention, in the shape extraction method of aspect 1 or aspect 2, the irradiation light is reflected by the second region and the first region toward the imaging unit, and the intensity with which the first region reflects the irradiation light toward the imaging unit and the intensity with which the second region reflects the irradiation light toward the imaging unit are different from each other.

[0013] In addition, according to a fourth aspect of the present invention, in the shape extraction method according to any one of the first to third aspects, the color of the reflective surface is a complementary color to the color of the tool.

[0014] In order to solve the above problem, a shape extraction method according to a fifth aspect of the present invention is a shape extraction method for photographing a tool including a first region and a second region adjacent to the first region, and extracting a shape of an extraction target part at least a part of which is located at a boundary between the first region and the second region, the shape extraction method including the first region and the second region includes a preparation step of arranging a light-absorbing unit having a light-absorbing surface such that the light-absorbing surface is located behind the tool when viewed from the photographing direction, an irradiation step of irradiating light toward the tool and the light-absorbing surface, a photographing step of photographing an image including the extraction target part by a photographing unit, and an extraction step of extracting the shape of the extraction target part from the image photographed in the photographing step, the irradiation light is reflected by the second region and the first region toward the photographing unit, and an intensity at which the first region reflects the irradiation light toward the photographing unit and an intensity at which the second region reflects the irradiation light toward the photographing unit are different from each other, and in the extraction step, the shape of the extraction target part is extracted from the image based on the light intensity in the image.

[0015] In order to solve the above problem, the wear detection method of aspect 6 of the present invention has a detection step of detecting wear of the extraction target part based on the shape of the extraction target part extracted by any one of the shape extraction methods of aspects 1 to 5.

[0016] In order to solve the above problem, a shape extraction device of aspect 7 of the present invention is a shape extraction device for photographing a tool including a first area and a second area adjacent to the first area, and extracting the shape of an extraction target part, at least a portion of which is located at the boundary between the first area and the second area, and comprises a reflecting unit having a reflective surface, an irradiation unit that irradiates light toward the tool and the reflecting surface, a photographing unit that photographs an image including the extraction target part, and an extraction unit that extracts the shape of the extraction target part from the image photographed by the photographing unit, wherein the reflecting surface has a color different from both the color of the irradiated light and the color of the tool, the irradiated light is reflected by the reflecting surface toward the second area, and the extraction unit extracts the shape of the extraction target part from the image based on the light intensity of a predetermined color in the image.

[0017] In order to solve the above problem, a shape extraction device according to an eighth aspect of the present invention is a shape extraction device for photographing a tool including a first region and a second region adjacent to the first region, and extracting a shape of an extraction target portion at least a portion of which is located at a boundary between the first region and the second region, the shape extraction device comprising: a light-absorbing unit having an absorbing surface and arranged so that the absorbing surface is located behind the tool when viewed from the photographing direction; an irradiation unit that irradiates irradiation light toward the tool and the light-absorbing surface; a photographing unit that photographs an image including the extraction target portion; and an extraction unit that extracts the shape of the extraction target portion from the image photographed by the photographing unit, wherein the irradiation light is reflected by the second region and the first region toward the photographing unit, and an intensity at which the first region reflects the irradiation light toward the photographing unit and an intensity at which the second region reflects the irradiation light toward the photographing unit are different from each other, and the extraction unit extracts the shape of the extraction target portion from the image based on the light intensity in the image.

[0018] In order to solve the above problem, a wear detection device according to a ninth aspect of the present invention is provided with a detection unit that detects wear of the extraction target part based on the shape of the extraction target part extracted by the shape extraction device of aspect 9 or aspect 10. Effect of the Invention

[0019] According to the above aspects of the present invention, it is possible to provide a shape extraction method and shape extraction device capable of clearly extracting the shape of an extraction target portion, and a wear detection method and wear detection device capable of accurately detecting wear on an extraction target portion. [Brief description of the drawings]

[0020] [Figure 1] FIG. 2 is a diagram illustrating a shape extraction device and a wear detection device according to an embodiment of the present invention. [Diagram 2] 2 is a diagram showing an extension portion shown in FIG. 1. FIG. [Diagram 3] FIG. 4 is a diagram showing an example of an image captured by a capturing section; [Figure 4] (a) is an image obtained by converting an image captured by the image capture unit into a grayscale image when the main surface is white or silvery white. (b) is an image obtained by extracting the shape of the blade tip based on the image (a). [Diagram 5] (a) is an image captured by the imaging unit when the main surface is light blue to light blue. (b) is an image obtained by extracting the R component from the image (a). [Figure 6] This is an image in which the shape of the cutting edge is extracted based on the image in FIG. 5(b). [Figure 7] 4 is a flowchart illustrating a shape extraction method and a wear detection method according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0021] Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

[0022] <Shape extraction device and wear detection device> FIG. 1 shows a state in which a tool 2 attached to a spindle 1 of a CNC (Computer Numerical Control) gear skiving machine (detailed illustration omitted) is observed by a shape extraction device 3 (wear detection device 4).

[0023] The shape extraction device 3 (wear detection device 4) may be disposed inside a processing device such as a gear skiving machine. More specifically, the shape extraction device 3 (wear detection device 4) may be provided inside the processing device at a position different from the position where the workpiece to be processed is attached to the processing device. However, the shape extraction device 3 and the wear detection device 4 may be provided separately from the processing device.

[0024] Hereinafter, the direction parallel to the central axis O of the spindle 1 and the tool 2 will be referred to as the axial direction Z. The direction from the tool 2 toward the spindle 1 along the axial direction Z will be referred to as the +Z direction or upward. The direction opposite to the +Z direction will be referred to as the -Z direction or downward. The direction perpendicular to the central axis O as viewed from the axial direction Z will be referred to as the radial direction. The direction approaching the central axis O along the radial direction will be referred to as the radial inner direction, and the direction moving away from the central axis O will be referred to as the radial outer direction. The direction going around the central axis O as viewed from the axial direction Z will be referred to as the circumferential direction.

[0025] The tool 2 according to this embodiment has a plurality of blades 20. Each blade 20 protrudes radially outward from an outer circumferential surface (a surface facing radially outward) of the tool 2. The plurality of blades 20 are arranged at intervals in the circumferential direction along the outer circumferential surface of the tool 2.

[0026] 1, in this embodiment, the multiple blades 20 are twisted at a predetermined angle in the axial direction Z. That is, each blade 20 extends such that its position in the radial direction also changes as its position in the axial direction Z changes. As a result, the side surface of each blade 20 (hereinafter referred to as the blade side surface 23) is inclined with respect to the axial direction Z. In other words, the normal direction of the blade side surface 23 is inclined with respect to the axial direction Z.

[0027] Each blade 20 has a blade bottom surface 22 located at the tip (lower end) of the blade 20 (see also FIG. 3). The normal direction of the blade bottom surface 22 is approximately parallel to the axial direction Z. The blade bottom surface 22 and the blade side surface 23 are adjacent to each other via the blade tip 21. In other words, the blade tip 21 is located at the boundary between the blade bottom surface 22 and the blade side surface 23. The blade bottom surface 22 and the blade side surface 23 face in different directions.

[0028] The shape extraction device 3 according to this embodiment is a device that extracts the shape of the cutting edge 21 from an image I obtained by capturing the blade 20. In other words, when a portion that is the subject of shape extraction by the shape extraction device 3 is referred to as an extraction target portion T, in this embodiment, the cutting edge 21 corresponds to the extraction target portion T. Furthermore, at least a portion of the cutting edge 21 that is the extraction target portion T is located at a boundary portion BD between the bottom blade surface 22 (first region AR1) and the blade side surface 23 (second region AR2). Furthermore, the wear detection device 4 according to this embodiment is a device that detects wear of the cutting edge 21 based on the shape of the cutting edge 21 extracted by the shape extraction device 3.

[0029] 1, the shape extraction device 3 according to this embodiment includes an imaging unit 31, an irradiation unit 32, a mirror 33, an extension unit 34, an extraction unit 35, and a base unit 36. The wear detection device 4 according to this embodiment includes the shape extraction device 3 and a detection unit 41.

[0030] The pedestal portion 36 is fixed, for example, in the above-mentioned processing apparatus. The pedestal portion 36 has a base portion 361, an upright portion 362, and a mirror fixing portion 363.

[0031] The base 361 extends so as to intersect with the axial direction Z. The standing portion 362 extends upward from one end of the base 361. The mirror fixing portion 363 is fixed to the upper surface of the base 361 in the vicinity of the standing portion 362. The mirror fixing portion 363 has a fixing surface 363a to which the mirror 33 is fixed. The fixing surface 363a is inclined at a predetermined angle θ with respect to the axial direction Z.

[0032] The extension portion 34 is fixed to the upper end portion of the standing portion 362. The extension portion 34 has a plate-like shape extending from the upper end portion of the standing portion 362 in parallel to the base portion 361.

[0033] 2, the extension portion 34 according to this embodiment has a notch 342 formed therein, which extends inward from the edge of the extension portion 34. The spindle 1 described above holds the tool 2 in a state in which the tool 2 is inserted into the notch 342. The spindle 1 may be configured to move so that the arbor portion of the tool 2 is inserted into the notch 342 and to rest at that position.

[0034] 1 and 2, the extension portion 34 has a main surface 341 facing downward (toward the mirror 33). The main surface 341 has a color different from both the color of the light (details will be described later) emitted by the irradiation portion 32 and the color of the tool 2. The main surface 341 may be painted in the different color with paint.

[0035] The photographing unit 31 has an imaging element capable of converting light into an electrical signal (image data), and is configured to be able to photograph an image I of the tool 2. An example of the photographing unit 31 is an industrial camera, but the type of the photographing unit 31 is not particularly limited as long as it is capable of photographing the image I, and can be changed as appropriate.

[0036] The imaging unit 31 according to this embodiment is fixed to the other end of the base 361 so that the lens 311 faces the mirror 33. In this specification, the direction along the optical axis of the imaging unit 31 is referred to as the "imaging direction S (of the imaging unit 31)."

[0037] In this embodiment, the optical axis (imaging direction S) of the imaging unit 31 is bent by the mirror 33, and the imaging unit 31 captures an image I of the tool 2 via the mirror 33. However, the position of the imaging unit 31 is not particularly limited as long as the image I of the tool 2 can be captured.

[0038] Fig. 3 is a diagram showing an example of image I captured by the imaging unit 31. As shown in Fig. 3, image I includes the blade bottom surface 22 which is the first region AR1, the blade side surface 23 which is the second region AR2, and the blade tip 21 which is the extraction target portion T. Note that the reason why the blade side surface 23 is reflected in image I is because the blade 20 is twisted and the blade side surface 23 is inclined with respect to the axial direction Z as described above.

[0039] In this embodiment, the above-described extension portion 34 is disposed such that the main surface 341 is located behind the tool 2 when viewed from the shooting direction S. Therefore, the main surface 341 is included in the image I. More specifically, the main surface 341 functions as the background of the tool 2 in the image I.

[0040] The irradiation unit 32 irradiates irradiation light toward the tool 2 and the main surface 341 via the mirror 33 (see FIG. 1). The irradiation light is, for example, white light. The irradiation unit 32 in the example of FIG. 1 is a ring-shaped light attached to the photographing unit 31 so as to surround the lens 311. As a result, the irradiation light according to this embodiment is irradiated toward the tool 2 and the main surface 341 along the photographing direction S of the photographing unit 31. However, the type and position of the irradiation unit 32 can be changed as appropriate as long as it is possible to irradiate irradiation light toward the tool 2 and the main surface 341.

[0041] The mirror 33 bends the shooting direction S of the photographing unit 31 toward the tool 2. The mirror 33 also reflects the irradiated light emitted from the irradiation unit 32 toward the tool 2 and the main surface 341. The mirror 33 also reflects the irradiated light reflected by the tool 2 (for example, the blade bottom surface 22 which is the first area AR1 and the blade side surface 23 which is the second area AR2) toward the photographing unit 31. The angle θ of the mirror 33 (fixed surface 363a) is appropriately set (for example, θ=45°) so that light is appropriately exchanged between the tool 2 and the main surface 341 and the photographing unit 31.

[0042] However, the shape extraction device 3 does not need to have the mirror 33. In this case, the photographing unit 31 may directly photograph the tool 2 from below, and the irradiating unit 32 may directly photograph the tool 2 and the main surface 341 from below.

[0043] The extraction unit 35 extracts the shape of the blade tip 21, which is the extraction target portion T, from the image I captured by the imaging unit 31. Hereinafter, the mechanism by which the extraction unit 35 extracts the shape of the extraction target portion T (blade tip 21) will be described in detail.

[0044] In general, the reflected light obtained by irradiating a surface with light includes specular reflected light and diffuse reflected light (diffuse reflected light). Specular reflected light is reflected light obtained by reflection where the angle of incidence and the angle of reflection are equal (specular reflection). Diffuse reflected light is reflected light obtained by reflection in any direction other than specular reflection. Specular reflected light has the highest intensity among all reflected lights. For example, the intensity of specular reflected light is higher than the intensity of diffuse reflected light.

[0045] When the main surface 341 of the extension 34 is white or white silver (i.e., the same color as the irradiated light), the shape of the cutting edge 21 detected by image processing such as binarization, ridge extraction, and Sobel filter tends to be unclear due to the influence of the blade side surface 23. Figure 4 is a diagram showing this situation.

[0046] The illumination light (white light) emitted from the illumination unit 32 is illuminated approximately perpendicularly to surface A (blade bottom surface 22) shown in Fig. 4(a). Therefore, of the illumination light reflected by the blade bottom surface 22, light that enters the imaging unit 31 is light that is specularly reflected by surface A (blade bottom surface 22). The imaging unit 31 images (captures) this specularly reflected light as surface A (blade bottom surface 22).

[0047] 4(b) is inclined with respect to the irradiated light (photographing direction S, axial direction Z). Therefore, the specular reflection light, which has the highest intensity among the reflected light from the irradiated light, does not head toward the photographing unit 31, and only a small portion of the diffuse reflection light enters the photographing unit 31.

[0048] Furthermore, surface B (blade side surface 23) is also irradiated with diffusely reflected light, which is generated when the irradiated light is diffusely reflected by surface C (principal surface 341 of extension portion 34). That is, principal surface 341 of extension portion 34 functions as a reflective surface that reflects (diffusely reflects) the irradiated light toward surface B (blade side surface 23). Extension portion 34, which has principal surface 341 as a reflective surface, is also referred to as a reflective portion.

[0049] The diffusely reflected light from surface C (principal surface 341) is further reflected by surface B (blade side surface 23). The diffusely reflected light contained in this reflected light travels toward the photographing unit 31. The photographing unit 31 photographs (captures) the light that is a combination of the diffusely reflected light from surface C (principal surface 341) and the very small portion of the diffusely reflected light described above as surface B (blade side surface 23). Note that surface A (blade bottom surface 22) is not irradiated with the diffusely reflected light that is generated when the irradiated light is diffusely reflected by surface C (principal surface 341).

[0050] In Fig. 4(a), surface A (blade bottom surface 22) is brighter than surface B (blade side surface 23) due to the effect of diffuse reflection from surface C (principal surface 341). Even if an attempt is made to extract the shape of cutting edge 21 by image processing such as ridge extraction under such circumstances, it becomes difficult to extract the shape of cutting edge 21, which is the extraction target portion T, at boundary BD between bottom surface 22 and blade side surface 23, as shown in Fig. 4(b).

[0051] In addition, when the extension 34 is removed and there is no main surface 341, surface B (blade side surface 23) becomes darker than surface A (blade bottom surface 22). This makes it easier to extract the shape of the cutting edge 21 at the boundary BD. However, when the extension 34 is removed, the chuck of the spindle 1 and the like are reflected on surface C (i.e., the background of the blade 20) in image I. Therefore, it becomes more difficult to extract the shape of the cutting edge 21 at the boundary between surface C and surface A (blade bottom surface 22) compared to when the main surface 341 is located on surface C. In addition, what is photographed as the main surface 341 in image I is the light reflected from the main surface 341 (through the mirror 33) toward the photographing unit 31.

[0052] To alleviate such difficulty in shape extraction, in the shape extraction device 3 (wear detection device 4) according to this embodiment, the main surface 341 (reflecting surface) of the extension portion 34 (reflecting portion) has a color different from both the color of the irradiated light (described in detail later) and the color of the tool 2. Then, the extraction unit 35 extracts the shape of the cutting edge 21 (extraction target portion T) from the image I based on the light intensity of a predetermined color in the image I.

[0053] An example in which the color of the main surface 341 (reflective surface) is light blue to light blue and the extraction unit 35 extracts the shape of the cutting edge 21 based on the light intensity of red in the image I will be described below.

[0054] In general, the blade bottom surface 22 and the blade side surface 23 of the tool 2 are a gray metallic color. Therefore, when decomposed into RGB components, the color of the blade bottom surface 22 and the blade side surface 23 contains all of the R component, the G component, and the B component.

[0055] Here, when the color of the main surface 341 (reflective surface) is bright blue to light blue, the light reflected by the main surface 341 toward the mirror 33, the blade side surface 23, etc. will also be bright blue to light blue. Therefore, when this reflected light is decomposed into RGB components, the B and G components are predominant, and the R component is extremely small compared to these two components.

[0056] When the tool 2 is photographed under such circumstances, an image I is obtained in which the blade bottom surface 22 contains the R component, but the main surface 341 (reflecting surface) contains almost no R component. As described above, what is photographed as the blade side surface 23 is a combination of a very small amount of diffusely reflected light of the irradiated light and light illuminated by the diffusely reflected light from the main surface 341 (reflecting surface). Therefore, the blade side surface 23 in image I also contains only a small amount of the R component, just like the main surface 341.

[0057] Therefore, by extracting the R component from the obtained image I using, for example, color separation of the color channel (i.e., extracting the red light intensity), the influence of the blade side surface 23 and main surface 341 can be reduced, and the shape of the blade bottom surface 22 (and the cutting edge 21 located at its contour) can be obtained.

[0058] FIG. 5 shows images obtained when shape extraction was actually performed using the above method. FIG. 5(a) is an image taken in the presence of a bright blue to light blue main surface 341 (reflective surface). FIG. 5(b) is an image obtained by extracting the R component from the image in FIG. 5(a). It can be seen that by extracting the R component (FIG. 5(b)) from the image (FIG. 5(a)) taken in the presence of a bright blue to light blue main surface 341 (reflective surface), the influence of the blade side surface 23 and main surface 341 is reduced, and the shape of the blade bottom surface 22 (and the blade tip 21 located at its contour) can be obtained.

[0059] By performing image processing such as contour extraction on the image obtained by extracting the R component in this manner, the extraction unit 35 can clearly extract the shape of the cutting edge 21 while reducing the influence of the blade side surface 23, as shown in Figure 6.

[0060] The above-mentioned color of the main surface 341 (reflecting surface) and the color components extracted by the extraction unit 35 are merely examples, and can be appropriately changed as long as the same effect can be obtained. For example, the color of the main surface 341 (reflecting surface) may be set to red, and the extraction unit 35 may perform a similar method using a color separation image of the B component or G component. If the color of the main surface 341 (reflecting surface) is different from both the color of the irradiated light and the color of the tool 2, it is expected that the same effect can be obtained by appropriately setting the color extracted by the extraction unit 35. In particular, if the color of the main surface 341 (reflecting surface) is set to the complementary color of the color of the tool 2, all of the diffusely reflected light from the main surface 341 toward the blade side surface 23 is absorbed by the blade side surface 23, making it easier to extract the shape of the blade bottom surface 22.

[0061] Also, color channels other than RGB (for example, CMYK, etc.) may be used. As long as it is possible to extract colors with different intensities between the blade side surface 23 (and the main surface 341) and the blade bottom surface 22, the color extracted by the extraction unit 35 and the specific extraction method are not limited and can be changed as appropriate.

[0062] Furthermore, the irradiated light does not have to be white as long as the same effect as described above can be obtained. For example, the irradiated light may have a frequency spectrum that lacks frequencies other than those necessary to obtain the above effect.

[0063] The detection unit 41 detects wear of the cutting edge 21 based on the shape of the cutting edge 21 (extraction target portion T) extracted by the shape extraction device 3 (extraction unit 35). The detection unit 41 may detect wear of the cutting edge 21 by, for example, storing the shape of the cutting edge 21 in an unworn state and comparing the stored shape with the shape of the cutting edge 21 extracted by the extraction unit 35.

[0064] The extraction unit 35 and the detection unit 41 can be realized by a computer such as a personal computer or a workstation. When the extraction unit 35 and the detection unit 41 are realized by a computer, the functions of the extraction unit 35 and the detection unit 41 are realized by a program for realizing those functions being executed by a CPU (Central Processing Unit) provided in the computer. In other words, the functions of the extraction unit 35 and the detection unit 41 are realized by the cooperation of software and hardware resources. The extraction unit 35 and the detection unit 41 may be realized by using hardware such as a Field-Programmable Gate Array (FPGA), a Large Scale Integration (LSI), or an Application Specific Integrated Circuit (ASIC).

[0065] <Shape extraction method and wear detection method> 7 is a flowchart showing a shape extraction method and a wear detection method according to this embodiment. Preparation step S11 to extraction step S14 correspond to the shape extraction method according to this embodiment, and preparation step S11 to detection step S15 correspond to the wear detection method according to this embodiment.

[0066] First, a preparation step S11 is performed. In the preparation step S11, a reflecting section having a reflecting surface is prepared. The color of the reflecting surface is as described above. At this time, the reflecting section is arranged so that the reflecting surface is located behind the tool 2 when viewed from the shooting direction S. For example, the extension section 34 (reflecting section) having a main surface 341 is arranged so that the main surface 341 is located behind the tool 2 when viewed from the shooting direction S. Such an arrangement relationship may be realized by moving the spindle 1 so that the arbor portion of the tool 2 is inserted into the notch portion 342 and stopping at that position.

[0067] After the preparation step S11 is performed, an irradiation step S12 is performed. In the irradiation step S12, irradiation light is irradiated toward the tool 2 and the reflecting surface. For example, the irradiation unit 32 of the shape extraction device 3 irradiates irradiation light toward the tool 2 and the main surface 341.

[0068] After the irradiation step S12, the photographing step S13 is performed. In the photographing step S13, an image I including the extraction target portion T is photographed by the photographing unit 31. For example, an image I including the blade tip 21 is photographed by the photographing unit 31 of the shape extraction device 3. At this time, due to the effect of diffuse reflection light from the main surface 341, the color component distribution of the first region AR1 (e.g., the blade bottom surface 22) and the color component distribution of the second region AR2 (e.g., the blade side surface 23) in the image I are different from each other.

[0069] After the photographing step S13, an extraction step S14 is performed. In the extraction step S14, the shape of the extraction target portion T is extracted from the image I photographed in the photographing step S13 based on the light intensity of a predetermined color. For example, the extraction unit 35 of the shape extraction device 3 extracts the shape of the blade edge 21 from the image I photographed by the photographing unit 31. The mechanism for extracting the shape of the blade edge 21 is as described above.

[0070] After the extraction step S14 is performed, a detection step S15 is performed. In the detection step S15, wear of the extraction target part T is detected based on the shape of the extraction target part T detected in the extraction step S14. For example, the detection unit 41 of the wear detection device 4 detects wear of the cutting edge 21 based on the shape of the cutting edge 21 extracted by the shape extraction device 3 (extraction unit 35).

[0071] <Summary> As described above, the shape extraction device 3 of this embodiment is a shape extraction device for photographing a tool 2 including a blade bottom surface 22 (first region AR1) and a blade side surface 23 (second region AR2) adjacent to the blade bottom surface 22, and extracting the shape of the cutting edge 21 (extraction target portion T) at least a part of which is located at the boundary portion BD between the blade bottom surface 22 and the blade side surface 23. The shape extraction device 3 is equipped with an extension portion 34 (reflection portion) having a main surface 341 (reflection surface), an irradiation portion 32 that irradiates irradiation light toward the tool 2 and the main surface 341, an imaging portion 31 that photographs an image I including the cutting edge 21, and an extraction portion 35 that extracts the shape of the cutting edge 21 from the image I photographed by the imaging portion 31, where the main surface 341 has a color different from both the color of the irradiation light and the color of the tool 2, the irradiation light is reflected by the main surface 341 toward the blade side surface 23, and the extraction portion 35 extracts the shape of the cutting edge 21 from the image I based on the light intensity of a predetermined color in the image I.

[0072] The shape extraction method according to the present embodiment is a shape extraction method for extracting the shape of the cutting edge 21 (extraction target portion T) at least a part of which is located at a boundary portion BD between the cutting bottom surface 22 and the cutting side surface 23 by photographing the tool 2 including the cutting bottom surface 22 (first region AR1) and the cutting side surface 23 (second region AR2) adjacent to the cutting bottom surface 22, and includes a preparation step S11 of preparing an extension portion 34 (reflection portion) having a main surface 341 (reflection surface), and an irradiation step of irradiating the tool 2 and the main surface 341 with irradiation light. The method includes a photographing step S12, a photographing step S13 in which an image I including the cutting edge 21 is photographed by the photographing unit 31, and an extraction step S14 in which the shape of the cutting edge 21 is extracted from the image I photographed in the photographing step S13, wherein the main surface 341 has a color different from both the color of the irradiated light and the color of the tool 2, and the irradiated light is reflected by the main surface 341 toward the blade side surface 23, and in the extraction step S14, the shape of the cutting edge 21 is extracted from the image I based on the light intensity of a predetermined color in the image I.

[0073] With this configuration, the difference between the color component distribution of the bottom blade surface 22 and the color component distribution of the blade side surface 23 in image I is utilized to reduce the influence of the blade side surface 23, making it possible to clearly extract the shape of the blade tip 21.

[0074] Further, the extension 34 is disposed so that the main surface 341 is located behind the tool 2 when viewed from the shooting direction S.

[0075] This configuration reduces the influence of something (such as the chuck of the spindle 1) located behind the tool 2, and allows for a clearer extraction of the shape of the cutting edge 21. However, the main surface 341, which is a reflecting surface, may be provided at a position other than behind the tool 2, as long as the shape of the cutting edge 21 can be extracted sufficiently clearly.

[0076] The irradiated light is reflected by the blade side surface 23 and the blade bottom surface 22 toward the photographing unit 31, and the intensity with which the blade bottom surface 22 reflects the irradiated light toward the photographing unit 31 differs from the intensity with which the blade side surface 23 reflects the irradiated light toward the photographing unit 31. Specifically, in this embodiment, the light that the blade bottom surface 22 reflects toward the photographing unit 31 is specularly reflected light, and the light that the blade side surface 23 reflects toward the photographing unit 31 is diffusely reflected light.

[0077] With this configuration, the difference between the color component distribution of the blade bottom surface 22 and the color component distribution of the blade side surface 23 in the image I can be made larger compared to when both intensities are approximately the same. This allows the shape of the blade edge 21 to be extracted more clearly. However, as long as the shape of the blade edge 21 can be extracted sufficiently clearly, the intensity with which the blade bottom surface 22 reflects the irradiated light toward the imaging unit 31 and the intensity with which the blade side surface 23 reflects the irradiated light toward the imaging unit 31 may be approximately the same.

[0078] In addition, the color of the main surface 341 may be a complementary color to the color of the tool 2 .

[0079] According to this configuration, the influence of the blade side surface 23 can be further reduced, and the shape of the blade tip 21 can be extracted more clearly.

[0080] Moreover, the wear detection device 4 according to this embodiment includes a detection unit 41 that detects wear of the cutting edge 21 based on the shape of the cutting edge 21 extracted by the above-described shape extraction device 3. Similarly, the wear detection method according to this embodiment includes a detection step S15 that detects wear of the cutting edge 21 based on the shape of the cutting edge 21 detected by the above-described wear detection method.

[0081] Since the shape of the cutting edge 21 is clearly extracted by the above-described shape extraction device 3 or wear detection method, wear detection can be performed based on the shape, thereby improving the accuracy of wear detection.

[0082] <Modification> The technical scope of the present invention is not limited to the above-described embodiment, and various modifications can be made without departing from the spirit of the present invention.

[0083] For example, in the above embodiment, the main surface 341 is a reflective surface that reflects (diffusely reflects) the irradiated light, but the main surface 341 may be an absorptive surface that absorbs the irradiated light. The extension 34 having the main surface 341 that is an absorptive surface is also referred to as a light absorbing portion. In this case, the main surface 341 may be black (especially carbon black or the like having high light absorption properties is preferable). Alternatively, the main surface 341 may be a color complementary to the color of the irradiated light.

[0084] When the main surface 341 is a light-absorbing surface, only a very small portion of the diffusely reflected light of the irradiated light is captured as the blade side surface 23. As described above, it is the specularly reflected light of the irradiated light that is captured as the blade bottom surface 22. Therefore, in the image I, the brightness (light intensity) of the blade bottom surface 22, which is the first region AR1, can be made to differ greatly from the brightness (light intensity) of the blade side surface 23, which is the second region AR2.

[0085] Therefore, the extraction unit 35 can extract the shape of the blade tip 21, which is the extraction target portion T, based on the light intensity in the image I, for example, by converting the obtained image I into a grayscale image.

[0086] Furthermore, the configuration of the extension portion 34 can be changed as appropriate so long as it has the main surface 341 that functions as a reflecting surface or a light absorbing surface. For example, the extension portion 34 does not have to be in the shape of a plate.

[0087] Also, the image I may be captured while rotating the tool 2 at a low speed or indexing rotation by the spindle 1 or the like. In this case, after extracting the shape of the cutting edge 21 and detecting wear of the cutting edge 21 of a certain blade 20, extracting the shape of the cutting edge 21 and detecting wear of the cutting edge 21 of the other blades 20 may be performed in sequence.

[0088] Furthermore, the first area AR1, the second area AR2, and the extraction target portion T are not limited to the blade bottom surface 22, the blade side surface 23, and the blade tip 21, and can be changed as appropriate.

[0089] In addition, within the scope of the spirit of the present invention, it is possible to replace the components in the above-described embodiments with well-known components as appropriate, and the above-described embodiments and variations may be combined as appropriate. [Explanation of symbols]

[0090] 2...Tool 3...Shape extraction device 4...Wear detection device 31...Photographing section 32...Irradiation section 34...Extension section (reflecting section, light absorbing section) 341...Main surface (reflecting surface, light absorbing surface) 35...Extraction section 41...Detection section AR1...First area AR2...Second area BD...Boundary section T...Extraction target section

Claims

1. 1. A shape extraction method for capturing an image of a tool including a first region and a second region adjacent to the first region, and extracting a shape of an extraction target portion, at least a portion of which is located in a boundary portion between the first region and the second region, comprising: A preparation step of preparing a reflecting portion having a reflecting surface; an irradiation step of irradiating the tool and the reflecting surface with irradiation light; a photographing step of photographing an image including the extraction target portion by a photographing unit; An extraction step of extracting a shape of the extraction target portion from the image captured in the photographing step, the reflective surface has a color different from both the color of the irradiated light and the color of the tool; The irradiation light is reflected by the reflecting surface toward the second area, In the extraction step, a shape of the extraction target portion is extracted from the image based on a light intensity of a predetermined color in the image. Shape extraction method.

2. In the preparation step, the reflecting portion is disposed such that the reflecting surface is located behind the tool when viewed from a photographing direction. The method of claim 1 .

3. the irradiation light is reflected by the second region and the first region toward the imaging unit, The intensity with which the first region reflects the irradiation light toward the imaging unit and the intensity with which the second region reflects the irradiation light toward the imaging unit are different from each other. The method of claim 1 .

4. The color of the reflective surface is complementary to the color of the tool. The method of claim 1 .

5. 1. A shape extraction method for capturing an image of a tool including a first region and a second region adjacent to the first region, and extracting a shape of an extraction target portion, at least a portion of which is located in a boundary portion between the first region and the second region, comprising: A preparation step of disposing a light absorbing unit having a light absorbing surface such that the light absorbing surface is located behind the tool when viewed from the shooting direction; an irradiation step of irradiating the tool and the light absorbing surface with irradiation light; a photographing step of photographing an image including the extraction target portion by a photographing unit; An extraction step of extracting a shape of the extraction target portion from the image captured in the photographing step, the irradiation light is reflected by the second region and the first region toward the imaging unit, an intensity at which the first region reflects the illumination light toward the imaging unit and an intensity at which the second region reflects the illumination light toward the imaging unit are different from each other; In the extraction step, a shape of the extraction target portion is extracted from the image based on light intensity in the image. Shape extraction method.

6. a detection step of detecting wear of the extraction target part based on the shape of the extraction target part extracted by the shape extraction method according to any one of claims 1 to 5, Wear detection methods.

7. 1. A shape extraction device for capturing an image of a tool including a first region and a second region adjacent to the first region, and extracting a shape of an extraction target portion, at least a portion of which is located in a boundary portion between the first region and the second region, comprising: A reflecting portion having a reflecting surface; An irradiation unit that irradiates irradiation light toward the tool and the reflecting surface; An image capturing unit that captures an image including the extraction target portion; An extraction unit that extracts a shape of the extraction target portion from the image captured by the imaging unit, the reflective surface has a color different from both the color of the irradiated light and the color of the tool; The irradiation light is reflected by the reflecting surface toward the second area, The extraction unit extracts a shape of the extraction target portion from the image based on a light intensity of a predetermined color in the image. Shape extractor.

8. 1. A shape extraction device for capturing an image of a tool including a first region and a second region adjacent to the first region, and extracting a shape of an extraction target portion, at least a portion of which is located in a boundary portion between the first region and the second region, comprising: A light absorbing unit having a light absorbing surface and disposed so that the light absorbing surface is located behind the tool when viewed from the shooting direction; an irradiation unit that irradiates irradiation light toward the tool and the light absorbing surface; An image capturing unit that captures an image including the extraction target portion; An extraction unit that extracts a shape of the extraction target portion from the image captured by the imaging unit, the irradiation light is reflected by the second region and the first region toward the imaging unit, an intensity at which the first region reflects the illumination light toward the imaging unit and an intensity at which the second region reflects the illumination light toward the imaging unit are different from each other; The extraction unit extracts a shape of the extraction target portion from the image based on light intensity in the image. Shape extractor.

9. a detection unit that detects wear of the extraction target portion based on the shape of the extraction target portion extracted by the shape extraction device according to claim 7 or 8, Wear detection device.

Images