Tool shape detection device and tool shape detection method

Abstract

To provide a tool shape detecting device capable of measuring a shape of a tool whose shape is unknown and detecting abnormality of the shape of the tool.SOLUTION: A tool shape detecting device 1 for a tool 12 installed on a main spindle 11 of a machine tool 2, includes a camera 22 for capturing the shape of the tool 12, a normal vector acquisition part 27 for calculating normal vectors at a plurality of points on an edge 13 of the tool captured by the camera 22, a normal vector comparison part 29 for comparing a first plurality of normal vectors and a second plurality of normal vectors acquired by the normal vector acquisition part 27, and a tool shape determination part 3 for determining that the shape of the tool 12 is changed when the first plurality of normal vector values and the second plurality of normal vector values are different from a prescribed threshold, as the comparison result by the normal vector comparison part 29.SELECTED DRAWING: Figure 2

Description

[Technical Field]

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

[0002] In recent years, with improvements in the motion performance of equipment (machine tools) in ultra-precision machining of workpieces, the weight of tool shape accuracy in machining accuracy has increased. The shape of the tool is measured, for example, using a tool shape measuring device as shown in Patent Document 1. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] International Publication No. 2020 / 090844 Summary of the Invention [Problem to be solved by the invention]

[0004] Generally, when measuring the shape of a tool, the measuring instrument needs to specify what shape of tool it is measuring. Typically, the tool shapes that can be defined (specified) are those of commercially available tools, such as ball end mills, radius end mills, and flat end mills.

[0005] Incidentally, there are times when workpieces are machined using tools with unknown or special shapes, and it is sometimes necessary to measure the shape of these unknown tools and detect any abnormalities in their shape.

[0006] Therefore, the present invention aims to provide a tool shape detection device and a tool shape detection method that can measure the shape of a tool whose shape is unknown and further detect abnormalities in the shape of the tool. [Means for solving the problem]

[0007] The invention described in claim 1 is a tool shape detection device installed on the spindle of a machine tool, comprising: a camera for photographing the shape of the tool; a specific vector acquisition unit for obtaining specific vectors from a plurality of points on the edge of the tool photographed by the camera; a specific vector comparison unit for comparing a first plurality of specific vectors obtained by the specific vector acquisition unit with a second plurality of specific vectors subsequently obtained by the specific vector acquisition unit; and a tool shape determination unit for determining that the shape of the tool has changed when, as a result of the comparison by the specific vector comparison unit, the values ​​of the first plurality of specific vectors and the values ​​of the second plurality of specific vectors differ by a predetermined threshold, wherein the specific vector is a normal vector, a tangent vector, or a specific inclination vector that is inclined by a certain angle with respect to the normal vector.

[0008] In the invention described in claim 2, the first plurality of specific vectors acquired by the specific vector acquisition unit are obtained by Normal Vector and an edge shape acquisition unit that acquires an image of the tool by the camera after the tool is used to acquire the edge of the tool, and a first plurality of vectors acquired by the specific vector acquisition unit. Normal Vector and a tool shape change amount acquisition unit that uses the edge of the tool determined by the edge shape acquisition unit to determine the amount of change in the shape of the tool after use relative to the shape of the tool before use.

[0009] The invention described in claim 3 is a shape detection device for a tool installed on a spindle of a machine tool, comprising: a camera that photographs the shape of the tool; a specific vector acquisition unit that photographs the tool with the camera before use of the tool and obtains specific vectors at multiple points on an edge of the tool; The tool has an edge shape acquisition unit that photographs the tool with the camera after use to determine the edge of the tool, and a tool shape change amount acquisition unit that uses the specific vector obtained by the specific vector acquisition unit and the edge of the tool obtained by the edge shape acquisition unit to determine the amount of change in the shape of the tool after use relative to the shape of the tool before use, wherein the specific vector is Normal Vector This is a tool shape detection device.

[0010] The invention of claim 4 is a tool shape detection device installed on a spindle of a machine tool, comprising: a camera that photographs the shape of the tool; an edge shape acquisition unit that photographs the tool with the camera before use of the tool and obtains an edge of the tool; a specific vector acquisition unit that photographs the tool with the camera after use of the tool and obtains specific vectors at multiple points on the edge of the tool; and a tool shape change amount acquisition unit that uses the specific vector obtained by the specific vector acquisition unit and the edge of the tool obtained by the edge shape acquisition unit to obtain an amount of change in the shape of the tool after use relative to the shape of the tool before use, and the specific vector is Normal Vector This is a tool shape detection device.

[0011] The invention described in claim 5 is a method for detecting the shape of a tool installed on the spindle of a machine tool, comprising: a specific vector acquisition step for determining specific vectors at multiple points on the edge of the tool photographed by a camera that photographs the shape of the tool; a specific vector comparison step for comparing a first plurality of specific vectors acquired in the specific vector acquisition step with a second plurality of specific vectors subsequently acquired in the specific vector acquisition step; and a tool shape determination step for determining that the shape of the tool has changed when, as a result of the comparison in the specific vector comparison step, the values ​​of the first plurality of specific vectors and the values ​​of the second plurality of specific vectors differ by more than a predetermined threshold value, wherein the specific vector is a normal vector, a tangent vector, or a specific inclination vector that is inclined at a certain angle with respect to the normal vector.

[0012] The invention described in claim 6 is such that the first plurality of specific vectors obtained in the specific vector acquisition step are before use of the tool. Normal Vector and after the tool is used, an edge shape acquisition step of acquiring an image of the tool by the camera to acquire the edge of the tool, and a first plurality of vectors acquired in the specific vector acquisition step. Normal VectorThe tool shape detection method according to claim 5, further comprising: a step of acquiring a tool shape change amount, which uses the edge of the tool determined in the step of acquiring the edge shape of the tool, to determine the amount of change in the shape of the tool after use relative to the shape of the tool before use. [Effects of the Invention]

[0013] The present invention provides a tool shape detection device and a tool shape detection method that can measure the shape of a tool whose shape is unknown and further detect abnormalities in the shape of the tool. [Brief explanation of the drawings]

[0014] [Figure 1] 1 is a diagram showing a schematic configuration of a tool shape detection device according to an embodiment of the present invention and a machine tool on which the tool shape detection device is installed; [Figure 2] 1 is a diagram showing a schematic configuration of a tool shape detection device according to an embodiment of the present invention; [Figure 3] 3A and 3B are diagrams showing edges and normal vectors of a tool obtained by the tool shape detection device according to the embodiment of the present invention. [Figure 4] FIG. 4 is an enlarged view showing a part of the edge of the tool and its normal vector in FIG. [Figure 5] 1A and 1B are diagrams showing edges and normal vectors of a tool before and after use, which are obtained by the tool shape detection device according to the embodiment of the present invention. [Figure 6] FIG. 6 is a diagram illustrating components of the normal vector shown in FIG. 5. [Figure 7] FIG. 6 is an enlarged view showing a part of the edge of the tool and its normal vector in FIG. [Figure 8] 4 is a flowchart showing the operation of the tool shape detection device according to the embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0015] The tool shape detection device 1 (tool shape abnormality detection device) 1 according to an embodiment of the present invention is used by being installed on a machine tool 2, for example, as shown in Figure 1.

[0016] The machine tool 2 has a table 16 and a gantry-shaped column 10 on the upper surface of the bed 18, and the spindle head 4 is supported on the cross rail 8 of the column 10 via saddles 6. The spindle head 4 supports the spindle 11.

[0017] Here, for the sake of explanation, we define a predetermined horizontal direction as the X direction (X-axis direction), another predetermined horizontal direction perpendicular to the X direction as the Y direction (Y-axis direction), and the vertical direction perpendicular to both the X and Y directions as the Z direction (Z-axis direction).

[0018] The table 16 is movable in the X-axis direction relative to the bed 18. The saddle 6 is movable in the Y-axis direction along the cross rail 8. The spindle head 4 is movable in the Z-axis direction relative to the saddle 6.

[0019] By moving these three axes, it is possible to move a tool (e.g., an end mill) 12 in three dimensions relative to a workpiece 14 placed on a table 16, thereby machining the workpiece 14. A tool shape detection device 1 is installed at the end of the table 16. A control device 20 is connected to the machine tool 2 and the tool shape detection device 1, and is able to control the machine tool 2 and the tool shape detection device 1. The control device 20 is configured with a CPU and memory, not shown.

[0020] Fig. 2 shows the measurement of the shape of a tool 12 by the tool shape detection device 1. The tool 12 is moved to the position shown in Fig. 2 using the three axes described above, and the shape of the tool 12 is measured. The tool shape detection device 1 includes a camera 22 and a lighting device 24, and the shape of the tool 12 is measured with the tool 12 positioned between the camera 22 and the lighting device 24 as shown in Fig. 2. Light from the lighting device 24 is applied to the back of the tool 12 to capture an image, so the shape of the tool 12 is captured as a shadow.

[0021] The camera 22 is equipped with a high-speed shutter, and is capable of taking still-image-like pictures even when the tool 12 is rotating at several thousand revolutions per minute. The camera 22 may also be equipped with a zoom lens, and the magnification ratio may be controlled by the control device 20. The spindle 11 is equipped with a rotation angle sensor (not shown), and the control device 20 can control the rotation speed, rotation angle positioning, and the like.

[0022] If the tool 12 rotates at a speed of 10,000 rpm or more, it is difficult to measure using only a high-speed shutter. In this case, the lighting device 24 is equipped with a strobe function. If a strobe with a short light emission time of a few microseconds is used, it is possible to measure the shape of the tool 12 even while it is rotating. The maximum rotation speed of the tool 12 can be set to about 120,000 rpm.

[0023] The tool 12 is used, for example, when forming the surface of a mold core or cavity by cutting. The cutting is performed, for example, to finish the surface of the mold core or cavity, and the cutting results in a mirror-like surface. The outer diameter of the end mill 12 is, for example, about 1 mm, and the rotation speed of the end mill 12 during cutting is about 60,000 rpm.

[0024] Incidentally, a still image of the maximum outer shape of the end mill 12 can be obtained by photographing the tool 12. This is because the location of this maximum outer shape is the cutting edge of the end mill 12, and the shape of the cutting edge affects the shape of the machined surface of the workpiece 14. Details of photographing the tool 12 are disclosed in International Publication No. 2020 / 090844.

[0025] The tool shape measuring device disclosed in International Publication No. 2020 / 090844 is used as the tool shape detecting device 1. The tool shape detecting device 1 detects the shape of a tool (e.g., a cutting tool such as a rotating end mill) 12 mounted on a spindle 11 of a machine tool (e.g., an ultra-precision machining center) 2.

[0026] 2, the tool shape detection device 1 is configured to include a control unit 25 and a camera (digital camera) 22 that captures an image of the shape of the tool 12. The control unit 25 is configured as, for example, a part of the control device 20, but the control unit 25 may also be provided separately from the control device 20.

[0027] Tool shape detection device 1 is a device that detects the shape of tool 12 attached to spindle 11 of machine tool (precision machining machine) 2. Tool 12 attached to spindle 11 of machine tool 2 rotates around a predetermined central axis C1. Tool 12 cuts workpiece 14 while rotating.

[0028] In addition, as shown in Figure 2, the tool shape detection device 1 is configured to include a normal vector acquisition unit (specific vector acquisition unit) 27, a normal vector comparison unit (specific vector comparison unit) 29, and a tool shape judgment unit 31.

[0029] The normal vector acquisition unit 27 is configured to obtain normal vectors at multiple points on the edge 13 of the tool 12 (the outer peripheral portion corresponding to the tip of the cutting edge; the edge of the maximum outer shape) photographed by the camera 22. The multiple points on the edge 13 of the tool 12 are indicated by reference characters Pn-1, Pn, Pn+1, Pn+2... in Figures 3, 4, etc. The multiple normal vectors are indicated by reference characters VPn-1, VPn, VPn+1, VPn+2... in Figures 3, 4, etc. The edge 13 of the tool 12 is obtained by the edge shape acquisition unit 33.

[0030] The multiple points on the edge 13 of the tool 12 are arranged in order in the extension direction of the edge 13, for example, at a constant very small distance from each other. The constant very small distance can be approximately the pixel pitch of the imaging element of the camera 22. The constant very small distance may be slightly larger than the pixel pitch.

[0031] Instead of obtaining the normal vector, the normal vector obtaining unit 27 may obtain a vector inclined at a certain angle with respect to the normal vector (specific inclination vector). In other words, a vector intersecting the normal vector at a certain angle may be obtained. For example, a vector of a tangent perpendicular to the normal vector (tangent vector) may be obtained. Here, the normal vector, specific inclination vector, and tangent vector are referred to as specific vectors. Furthermore, the normal vector obtaining unit may obtain the inclination of the tangent to the edge 13 or the inclination of the normal at each of a plurality of points on the edge 13 of the tool 12.

[0032] The normal vector comparison unit 29 is configured to compare the first plurality of normal vectors acquired by the normal vector acquisition unit 27 with the second plurality of normal vectors subsequently acquired by the normal vector acquisition unit 27. The first plurality of normal vectors are, for example, the plurality of normal vectors before cutting the workpiece 14 with the tool 12, and the second plurality of normal vectors are, for example, the plurality of normal vectors after cutting the workpiece 14 with the tool 12.

[0033] The tool shape determination unit 31 is configured to determine that the shape of the tool 12 has changed when, as a result of the comparison in the normal vector comparison unit 29, the values ​​of the first plurality of normal vectors and the values ​​of the second plurality of normal vectors differ by more than a predetermined threshold value.

[0034] Here, the acquisition of the normal vectors VPn-1, VPn, VPn+1, VPn+2, . . . by the normal vector acquisition unit 27 will be described in more detail with reference to FIG. 4, taking the normal vector VPn+1 as an example.

[0035] First, a line segment LAn is obtained, connecting two adjacent points Pn and Pn+1 on the edge 13 of the tool 12. Also, a line segment LAn+1 is obtained, connecting two adjacent points Pn+1 and Pn+2 on the edge 13 of the tool 12. Next, a normal vector VPn+1 is obtained, the intersection angle with the line segment LAn being αn+1 and the intersection angle with the line segment LAn being βn+1, and the normal vector VPn+1 having its starting point at point Pn+1. The normal vector VPn+1 points toward the center of the tool 12 (toward the rotation central axis C1). Also, the intersection angle αn+1 and the intersection angle βn+1 are equal to each other.

[0036] Other normal vectors can be found in the same way as the normal vector VPn+1. Note that multiple normal vectors VPn-1, VPn, VPn+1, VPn+2, etc., have equal absolute values ​​(scalar quantities). Multiple normal vectors VPn-1, VPn, VPn+1, VPn+2, etc., are, for example, unit vectors.

[0037] Note that a half-ray may be found from multiple points (such as point Pn) using the least squares method, instead of line segment LAn, to determine normal vector VPn, etc. That is, in FIG. 4, instead of line segment LAn+1, a first half-ray extending diagonally upward (toward point Pn+1) from point Pn+2 is found. This half-ray is found by the least squares method from multiple points (for example, two points Pn+1 and Pn). Similarly, a second half-ray extending diagonally downward (toward point Pn+3) from point Pn+2 is found. Next, a normal vector VPn+2 is found, starting from point Pn+2, with an intersection angle αn+2 with the first half-ray and an intersection angle βn+2 with the second half-ray. Note that the intersection angles αn+2 and βn+2 are equal. Other normal vectors are found in a similar manner using half-rays.

[0038] Next, the comparison between the first set of normal vectors and the second set of normal vectors in the normal vector comparison unit 29 will be explained in more detail with reference to Figures 5 and 6. The solid lines in Figure 5 represent the edges 13 of the tool (tool before use) 12 related to the first set of normal vectors. The dashed lines in Figure 5 represent the edges 13a of the tool (tool after use) 12 related to the second set of normal vectors. Note that in Figure 5, a portion of the dashed line 13a is separated from the solid line 13, while other parts of the dashed line 13a overlap with the solid line 13.

[0039] The first set of normal vectors are denoted by reference symbols VPn, VPn+1, VPn+2, VPn+3, VPn+4, etc., and the second set of normal vectors are denoted by reference symbols VQn, VQn+1, VQn+2, VQn+3, VQn+4, etc. The second set of normal vectors VQn, VQn+1, VQn+2, VQn+3, VQn+4, etc., are determined in the same way as the first set of normal vectors, and are, like the first set of normal vectors, for example, unit vectors.

[0040] FIG. 6 shows the components of each of the first plurality of normal vectors VPn, VPn+1, VPn+2, VPn+3, VPn+4, etc., and the components of each of the second plurality of normal vectors VQn, VQn+1, VQn+2, VQn+3, VQn+4, etc.

[0041] For example, the components of the normal vector VPn+1 are represented by (Pan+1, Pbn+1), and the components of the normal vector VQn+1 are represented by (Qan+1, Qbn+1).

[0042] The normal vector comparison unit 29 compares the value of Pan+1 / Pbn+1 (the direction of the normal vector) with the value of Qan+1 / Qbn+1. The normal vector comparison unit 29 also performs similar comparisons for other normal vectors. For example, it compares the value of Pan / Pbn with the value of Qan / Qbn, and also compares the value of Pan+2 / Pbn+2 with the value of Qan+2 / Qbn+2.

[0043] As a result of the above comparison, the value (Pan+1 / Pbn+1) of the normal vector VPn+1 of point Pn+1 and the value (Qan+1 / Qbn+1) of the normal vector VQn+1 of point Qn+1 shown in Figure 5 match. In other words, the difference between the value (Pan+1 / Pbn+1) of the normal vector VPn+1 of point Pn+1 and the value (Qan+1 / Qbn+1) of the normal vector VQn+1 of point Qn+1 is smaller than a predetermined threshold.

[0044] On the other hand, the value of the normal vector VPn+2 of point Pn+2 (Pan+2 / Pbn+2) and the value of the normal vector VQn+2 of point Qn+2 (Qan+2 / Qbn+2) are different from each other. That is, the difference between the value of the normal vector VPn+2 (Pan+2 / Pbn+2) and the value of the normal vector VQn+2 of point Qn+2 (Qan+2 / Qbn+2) is greater than a predetermined threshold.

[0045] A similar comparison reveals that the values ​​of the normal vectors VPn+3 and VQn+3 are different, and the values ​​of the normal vectors VPn+4 and VQn+4 are different. Furthermore, the values ​​of the normal vectors VPn+5 and VQn+5 are different, while the values ​​of the normal vectors VPn+6 and VQn+6 are identical.

[0046] The tool shape determination unit 31 then determines that the shape of the tool 12 has changed in the area from point Pn+2 (point Qn+2) to point Pn+5 (point Qn+5).

[0047] Next, we will explain how to detect the amount of wear on tool 12.

[0048] As described above, the first plurality of normal vectors acquired by the normal vector acquisition unit 27 are normal vectors before use of the tool 12. The tool shape detection device 1 is also configured to include an edge shape acquisition unit 33 and a tool shape change amount acquisition unit 35.

[0049] The edge shape acquisition unit 33 is configured to take a picture of the tool 12 with the camera 22 after the tool 12 has been used to determine the edges 13 of the tool 12. The normal vector acquisition unit 27, as described above, uses multiple points on the edges 13 of the tool 12 and the edges 13 themselves to determine the normal vector.

[0050] The tool shape change acquisition unit 35 uses the first set of normal vectors obtained by the normal vector acquisition unit 27 and the edges 13 of the tool 12 obtained by the edge shape acquisition unit 33 to determine the amount of change in the shape of the tool 12 after use relative to the shape of the tool 12 before use. The amount of change in the shape of the tool 12 can include the amount of wear on the tool 12, the amount of damage to the damaged parts of the tool 12, and so on.

[0051] Here, the process of obtaining the amount of change in the shape of the tool 12 by the tool shape change amount obtaining unit 35 will be described in more detail with reference to FIG.

[0052] The method for determining the change in the shape of tool 12 will be explained using point Pn+2 as an example. First, find the normal vector VPn+2 of point Pn+2. Next, find the line that passes through point Pn+2 and whose slope coincides with the normal vector VPn+2 (the equation of the line that starts at point Pn+2 and contains the normal vector VPn+2).

[0053] Next, we find the intersection point Qm+2 between the above line and edge 13a, and then determine the length of the line segment Ln+2 connecting point Pn+2 and intersection point Qm+2. The value of the length of this line segment Ln+2 represents the change in the shape of tool 12 at point Pn+2. Similarly, we determine the change in the shape of tool 12 for other points such as point Pn+3.

[0054] Next, the operation of the tool shape detecting device 1 will be described with reference to FIG.

[0055] In the initial state, the tool 12 is rotating, and the shape of the tool 12 can be measured by the tool shape detection device 1 as shown in FIG.

[0056] In the above initial state, under the control of the control device 20 (control unit 25), the tool 12 before use is photographed by the camera 22 (S1), and the normal vector of the tool 12 photographed in step S1 is obtained by the normal vector acquisition unit 27 (S3).

[0057] Next, the tool 12 is used to perform a predetermined cutting process on the workpiece 14 (S5), and the tool 12 after use is photographed by the camera 22 in step S55 (S7).

[0058] Next, the edge shape acquisition unit 33 obtains the edge 13 of the tool 12 photographed in step S7, and the normal vector acquisition unit 27 obtains the normal vector of the tool 12 photographed in step S7 (S9).

[0059] Next, the normal vector of the tool 12 obtained in step S3 is compared with the normal vector of the tool 12 obtained in step S9 by the normal vector comparison unit 29 (S11). Next, based on the comparison result in step S11, the tool shape determination unit 31 determines whether the shape of the tool 12 has changed (S13).

[0060] Next, the tool shape change amount obtaining unit 35 obtains the amount of change in the shape of the tool 12 from the normal vector of the tool 12 obtained in step S3 and the edge of the tool 12 obtained in step S9 (S15).

[0061] In the tool shape detection device 1, the normal vector acquisition unit 27 obtains normal vectors for multiple points on the edge 13 of the tool 12, and the normal vector comparison unit 29 compares the first set of normal vectors with the second set of normal vectors. When the values ​​of the first set of normal vectors and the second set of normal vectors differ by more than a predetermined threshold, the tool shape determination unit 31 determines that the shape of the tool 12 has changed.

[0062] This makes it possible to measure the shape of the tool 12, the shape of which is unknown, and further to detect abnormalities in the shape of the tool 12. Furthermore, by comparing the normal vectors, it is determined whether the shape of the tool 12 has changed before and after use, so that changes in the shape of the tool 12 can be detected more accurately than when simply photographing the tool 12 before and after use and comparing the shapes of the tool 12.

[0063] Incidentally, the above-mentioned tool shape detection device 1 may also be understood as a tool shape detection device installed on the spindle of a machine tool, having a camera that photographs the shape of the tool, a normal vector acquisition unit that photographs the tool with the camera before use and obtains normal vectors at multiple points on the edge of the tool, an edge shape acquisition unit that photographs the tool with the camera after use and obtains the edge of the tool, and a tool shape change amount acquisition unit that uses the normal vector obtained by the normal vector acquisition unit and the edge of the tool obtained by the edge shape acquisition unit to obtain the amount of change in the shape of the tool after use relative to the shape of the tool before use.

[0064] In this tool shape detection device 1, the normal vector acquisition unit 27 determines normal vectors at multiple points on the edge 13 of the tool 12 before use, and the edge shape acquisition unit 33 determines the edge 13a of the tool 12 after use. Then, using the normal vectors determined by the normal vector acquisition unit 27 and the edge 13a of the tool 12 determined by the edge shape acquisition unit 33, the tool shape change amount acquisition unit 35 determines the amount of change in the shape of the tool 12 after use relative to the shape of the tool 12 before use.

[0065] This makes it possible to accurately detect the amount of damage or wear caused by use and the part of the tool 12 where wear or the like has occurred.

[0066] Furthermore, the above-mentioned tool shape detection device 1 may also be understood as a tool shape detection device installed on the spindle of a machine tool, having a camera that photographs the shape of the tool, an edge shape acquisition unit that photographs the tool with the camera before use and determines the edge of the tool, a normal vector acquisition unit that photographs the tool with the camera after use and determines normal vectors at multiple points on the edge of the tool, and a tool shape change amount acquisition unit that uses the normal vectors determined by the normal vector acquisition unit and the edge of the tool determined by the edge shape acquisition unit to determine the amount of change in the shape of the tool after use relative to the shape of the tool before use.

[0067] The above description may also be understood as a method for detecting the shape of a tool.

[0068] In other words, this may be understood as a method for detecting the shape of a tool installed on the spindle of a machine tool, comprising: a normal vector acquisition step for determining normal vectors at multiple points on the edge of the tool photographed by a camera that photographs the shape of the tool; a normal vector comparison step for comparing a first plurality of normal vectors acquired in the normal vector acquisition step with a second plurality of normal vectors subsequently acquired in the normal vector acquisition step; and a tool shape determination step for determining that the shape of the tool has changed when, as a result of the comparison in the normal vector comparison step, the values ​​of the first plurality of normal vectors and the values ​​of the second plurality of normal vectors differ by more than a predetermined threshold value.

[0069] Furthermore, the first plurality of normal vectors acquired in the normal vector acquisition step are normal vectors of the tool before use, and the method may be understood as a tool shape detection method described in claim 5, which includes an edge shape acquisition step in which, after use of the tool, the tool is photographed with the camera to obtain the edges of the tool, and a tool shape change amount acquisition step in which, using the first plurality of normal vectors obtained in the normal vector acquisition step and the edges of the tool obtained in the edge shape acquisition step, the amount of change in the shape of the tool after use relative to the shape of the tool before use is obtained.

[0070] Although the present embodiment has been described above, the present embodiment is not limited to this, and various modifications are possible within the scope of the gist of the present embodiment. [Explanation of symbols]

[0071] 1. Tool shape detection device 2 Machine tools 11 Spindle 12 Tools 13, 13a Tool edge 22 cameras 27 Normal vector acquisition unit 29 Normal Vector Comparison Section 31 Tool shape judgment section 33 Edge shape acquisition unit 35 Tool shape change amount acquisition unit

Claims

1. A shape detection device for a tool installed on a spindle of a machine tool, a camera for photographing the shape of the tool; a specific vector acquisition unit that obtains specific vectors at a plurality of points on the edge of the tool photographed by the camera; a specific vector comparison unit that compares a first plurality of specific vectors acquired by the specific vector acquisition unit with a second plurality of specific vectors subsequently acquired by the specific vector acquisition unit; a tool shape determination unit that determines that the shape of the tool has changed when, as a result of the comparison by the specific vector comparison unit, the values ​​of the first plurality of specific vectors and the values ​​of the second plurality of specific vectors differ by more than a predetermined threshold value; wherein the specific vector is a normal vector, a tangent vector, or a specific tilt vector that is tilted at a certain angle with respect to the normal vector.

2. the first plurality of specific vectors acquired by the specific vector acquisition unit are normal vectors of the tool before use, an edge shape acquisition unit that acquires an image of the tool with the camera after the tool is used to determine the edge of the tool; a tool shape change amount acquisition unit that acquires a change amount of the shape of the tool after use relative to the shape of the tool before use, using a first plurality of normal vectors acquired by the specific vector acquisition unit and the edge of the tool acquired by the edge shape acquisition unit; The tool shape detection device according to claim 1 , further comprising:

3. A shape detection device for a tool installed on a spindle of a machine tool, a camera for photographing the shape of the tool; a specific vector acquisition unit that acquires specific vectors at a plurality of points on an edge of the tool by photographing the tool with the camera before the tool is used; an edge shape acquisition unit that acquires an image of the tool with the camera after the tool is used to determine the edge of the tool; a tool shape change amount acquisition unit that acquires a change amount of the shape of the tool after use relative to the shape of the tool before use, using the specific vector acquired by the specific vector acquisition unit and the edge of the tool acquired by the edge shape acquisition unit; wherein the specific vector is a normal vector.

4. A shape detection device for a tool installed on a spindle of a machine tool, a camera for photographing the shape of the tool; an edge shape acquisition unit that acquires an image of the tool with the camera before use of the tool to determine an edge of the tool; a specific vector acquisition unit that, after use of the tool, photographs the tool with the camera and obtains specific vectors at a plurality of points on an edge of the tool; a tool shape change amount acquisition unit that acquires a change amount of the shape of the tool after use relative to the shape of the tool before use, using the specific vector acquired by the specific vector acquisition unit and the edge of the tool acquired by the edge shape acquisition unit; wherein the specific vector is a normal vector.

5. A method for detecting the shape of a tool installed on a spindle of a machine tool, comprising: a specific vector acquisition step of obtaining specific vectors at a plurality of points on the edge of the tool photographed by a camera that photographs the shape of the tool; a specific vector comparison step of comparing a first plurality of specific vectors acquired in the specific vector acquisition step with a second plurality of specific vectors acquired next in the specific vector acquisition step; a tool shape determination step of determining that the shape of the tool has changed when the values ​​of the first plurality of specific vectors and the values ​​of the second plurality of specific vectors differ by more than a predetermined threshold value as a result of the comparison in the specific vector comparison step; wherein the specific vector is a normal vector, a tangent vector, or a specific inclined vector inclined at a certain angle with respect to the normal vector.

6. the first plurality of specific vectors acquired in the specific vector acquisition step are normal vectors of the tool before use, an edge shape acquisition step of photographing the tool with the camera after use of the tool to obtain an edge of the tool; a tool shape change amount acquisition step of acquiring a change amount of the shape of the tool after use relative to the shape of the tool before use, using a first plurality of normal vectors acquired in the specific vector acquisition step and the edge of the tool acquired in the edge shape acquisition step; The tool shape detection method according to claim 5, further comprising:

Images