JP7253088B2 - Image processing device, machine tool and image processing method - Google Patents

Description

The present invention relates to an image processing device, a machine tool, and an image processing method.

When machining a workpiece using a machine tool, it is possible to accurately continue machining the workpiece by grasping the shape and condition of the tool. Specifically, in machine tool machining, it is essential to check the interference of tools and the like, so it is important to accurately grasp the shape of the tool. Also, by grasping the state of the tool, it is possible to achieve high-precision machining. However, the current situation is that grasping the shape and condition of the tool requires special work and time.

Patent Literature 1 discloses a technique for determining when to replace an ultrasonic welding tool using an image. Further, Japanese Patent Laid-Open No. 2002-200002 discloses a technique for managing tools of a grinding apparatus using an image. Patent Literature 3 discloses a technique for capturing the tip of a tool in the field of view in response to changes in the tool of a machine tool. Furthermore, Patent Document 4 discloses a technique for inspecting a tool attached to a machine tool using an image.

JP 2010-207837 A JP 2001-269844 A JP-A-9-323240 JP 2016-218550 A

However, with any of the above patent documents, it is difficult to accurately grasp the shape and condition of the tool.

The image processing apparatus of the present disclosure includes a receiving unit that receives a plurality of image data of the tool captured by an imaging unit attached to a machine tool that processes a work using the tool, and a plurality of image data received by the receiving unit. a point cloud data forming unit that forms point cloud data, which is profile information related to the profile of the tool, from the point cloud data; Prepare.
In addition to the image processing apparatus, the present disclosure provides a machine tool, an image processing method, and the like.

These general and specific aspects may be implemented by systems, methods and computer programs, and combinations thereof.

According to the present invention, it is possible to accurately grasp the shape and condition of the tool.
By being able to accurately grasp the shape and condition of the tool, the work load on the machine tool can be reduced. Moreover, the improvement of processing precision can be aimed at.

It is an external view of a machine tool. FIG. 4 is a diagram for explaining the positional relationship among a tool, an imaging section, and an illumination section in the machine tool according to Embodiment 1; 1 is a block diagram showing the configuration of an image processing apparatus according to Embodiment 1; FIG. It is an example of the image data imaged with the machine tool. It is an example of point cloud data formed by an image processing device. It is an example of profile image data formed by an image processing apparatus. 5 is a flowchart for explaining processing for registering point cloud data according to Embodiment 1. FIG. FIG. 10 is a flowchart for explaining point cloud data update processing according to the second embodiment; FIG. FIG. 11 is a sequence diagram illustrating profile information update processing;

An image processing apparatus, a machine tool, and an image processing method according to each embodiment will be described below with reference to the drawings. In the following description, the same reference numerals are assigned to the same configurations, and descriptions thereof are omitted.

The "machine tool" described below is also referred to as a machine tool, and uses a tool to process a work such as a metal to be processed into a desired shape by cutting, grinding, or the like.

The "image processing device" of this embodiment is a device that processes an image captured within the machine tool. Furthermore, for example, it is used for management of tools used for machining a work in a machine tool. Specifically, the image processing device acquires image data of the tool. Based on the image data of the tool, the image processing device forms point cloud data and profile image data representing the two-dimensional or three-dimensional outline of the tool. Alternatively, the image processing device determines the blade portion of the tool and the profile such as the outer shape based on the image data of the tool. Alternatively, the image processing device determines the boundary between the blade portion and the shank portion of the tool based on the image data of the tool. The image processing apparatus can use these to control the position of the tool and manage the tool during machining.

"Point cloud data" is profile information for specifying a profile such as the outer shape of a tool, and will be described as point cloud data formed by a plurality of points extracted from the contour of the image data of the tool. In addition, the point cloud data includes tool profile information such as information specifying the region of the cutting edge and shank of the tool, the length of the tool, the outer diameter of the tool, and the outer diameter corresponding to the position in the longitudinal direction of the tool. Information about each part of the tool may be associated, including diameter information and the like.

"Profile image data" represents the two-dimensional or three-dimensional outline and contour of the tool. For example, two-dimensional profile image data is a projection drawing representing the contour of a tool. Also, the three-dimensional profile image data is, for example, three-dimensional image data representing the outer shape of the tool.

[Embodiment 1]
〈Machine tools〉
The machine tool 1 has an outer shape as shown in FIG. 1, for example, and can machine a work arranged in a machining area 200.

The configuration of the machine tool will be described with reference to FIGS. 1 and 2. FIG. The machine tool 1 includes a tool 11 attached to a spindle and used for machining a workpiece, a holder 12 capable of holding a workpiece to be machined and capable of holding the workpiece by being driven by a drive mechanism, and a tool. and an illumination unit 14 capable of emitting light. As shown in FIG. 1, the machine tool 1 includes a cover 201 that isolates a machining area 200 from the outside of the machine tool 1 . It has a door 203 capable of opening and closing an opening 202 for taking out to the outside. A control panel 15 is provided for operating the machine tool 1 such as machining.

Furthermore, the machine tool 1 is connected to an image processing device 10 which will be described later with reference to FIG. In this embodiment, the machine tool 1 and the image processing device 10 are separate devices and are connected via a wired cable. However, it is not limited to this form, and the image processing device 10 may also be incorporated inside the machine tool 1 .

The tool 11 includes, as shown in FIG. 2, a blade portion 111 which is a portion used for machining a workpiece, and a shank portion 112 which is a portion held by a holder 114 of a spindle 113 . Moreover, the machine tool 1 stores a plurality of types of tools 11 in a magazine 16, which is a storage portion shown in FIG. The machine tool 1 can implement a plurality of different types of machining by causing the spindle 113 to hold a tool 11 selected from a plurality of tools. The configuration shown in FIG. 2 is located in the dashed line portion in FIG.

The spindle 113 holds the tool 11 used for machining, and is configured to be able to rotate, move, etc. by a control mechanism. As a result, the machine tool 1 can machine the target work.

The imaging unit 13 is, for example, a camera equipped with an imaging device such as a CCD or CMOS. The imaging unit 13 can image the tool 11 attached to the tool spindle 113 at a predetermined timing. The imaging unit 13 of the present embodiment takes the tool 11 held by the tool spindle 113 as an imaging target. Note that the imaging unit 13 may capture the workpiece held by the workpiece holder of the workpiece spindle.

The imaging unit 13 images the tool 11 attached to the tool spindle 113 from a plurality of different directions. Specifically, the main shaft 113 rotates the held tool 11 at a predetermined pitch, and the imaging unit 13 performs imaging while the rotation of the tool 11 is stopped. By capturing images of the tool 11 at predetermined pitch intervals in this way, even when the imaging unit 13 is fixed, the tool 11 can be imaged from a plurality of different directions. In addition, when the tool 11 cannot fit in one image, the spindle 113 may move the tool 11 along the rotation axis.

The imaging unit 13 rotates the tool spindle 113 at predetermined intervals and images the tool 11 at each interval to obtain a plurality of images. In other words, an image of the entire tool 11 can be obtained by dividing it into a plurality of images. Therefore, the imaging unit 13 obtains a plurality of image data of one tool 11 viewed from different directions. Note that the imaging unit 13 outputs the obtained image data to the image processing device 10 .

As described above, the machine tool 1 can obtain image data of the tool 11 from a plurality of directions by imaging the tool 11 attached to the tool spindle 113 of the machine tool 1 with the imaging unit 13 . At this time, it is preferable that the imaging unit 13 acquires a rotation image of the tool 11 at a predetermined pitch. For example, the imaging unit 13 captures image data each time the tool 11 is rotated by 18° with respect to the rotation axis of the main shaft 113 , so that for each tool 11 , Multiple images from different directions can be obtained. At this time, the machine tool 1 provides information about the rotation angle of the tool 11 to the image processing device 10 together with the image data. Thereby, the image processing device 10 can generate the point cloud data of the three-dimensional information of the tool 11 .

For example, the imaging unit 13 images the tool 11 before the new tool 11 is stored in the magazine 16 that is the storage unit of the machine tool 1 . As a result, the machine tool 1 can easily generate the model data of the tool 11 for interference check in the simulation performed before machining, and can efficiently perform the interference check before machining.

As shown in FIG. 2, the illumination unit 14 is arranged such that the light emitting surface 141 faces the light receiving surface 131 of the imaging unit 13, and the light emitted from the light emitting surface 141 and passing around the tool is received. Incident on surface 131 . Therefore, this illumination unit 14 is a transmitted illumination. The use of the illumination unit 14, which is transmitted illumination, makes it possible to obtain a projected image from which the details of the outline of the tool 11 can be grasped. As a result, accurate point cloud data of the tool 11 can be obtained in the image processing device 10 that processes the projected image.

Note that the imaging unit 13 is arranged in an imaging storage unit having a shutter for partitioning the machining area except when photographing the tool 11 . When photographing the tool 11, the shutter of the imaging storage part is opened, the tool 11 attached to the main shaft in front of the imaging part 13 is moved by moving the main shaft, and the tool 11 is photographed. It is preferable to calibrate the imaging unit 13 before photographing the tool 11 . An object whose distance from the imaging unit 13 is known in advance is photographed, and it is confirmed from the photographed image whether the distance to the object is as set as the set value, and if different, the position of the imaging unit 13 is adjusted. The object whose distance is known in advance may be the rear surface of the shutter. Further, it may be detected that a foreign object such as dust is attached to a lens or glass that is a light transmission member of the imaging unit 13 or a CMOS that is a light receiving element. This is because there is a possibility that if a foreign substance adheres and is reflected in the captured image, it may be determined as chips in discrimination by AI. When the imaging unit 13 detects a foreign object, the control unit applies vibration to the light receiving element and the light transmitting member using a vibration mechanism, and performs control to remove the foreign object from the light transmitting member and the light receiving element. It is possible to obtain an image with reduced reflection. This makes it possible to accurately determine chips.

The positions of the imaging unit 13 and the tool 11 differ depending on the purpose of imaging. In other words, the tool 11 is set at a predetermined position so that the portion of the tool 11 whose profile information is to be generated is within the depth of field when the image is captured by the imaging unit 13 . For example, comparing the "contour measurement" of the tool 11 and the "surface wear measurement", in the case of the contour measurement of the tool 11, the contour portion is positioned on the working distance (WD) of the optical axis. On the other hand, when measuring the surface wear of the tool 11, the central portion of the diameter of the tool 11 is positioned on the optical axis and the outer surface of the tool 11 is positioned on the working distance (WD) of the optical axis. A desired image can be obtained by capturing an image in this manner.

<Image processing device>
An example of the image processing apparatus 10 according to the embodiment will be described with reference to FIG. The image processing apparatus 10 includes an arithmetic unit 100 , a storage unit 110 , an input unit 120 , an output unit 130 and a communication unit 140 . This image processing apparatus 10 is, for example, an information processing apparatus such as a personal computer or a tablet terminal.

The calculation unit 100 is a controller that controls the entire image processing apparatus 10 . For example, the calculation unit 100 reads out and executes the control program P stored in the storage unit 110 to obtain the receiving unit 101, the point cloud data forming unit 102, the profile image forming unit 103, the setting unit 104, and the storage processing unit 105. And the processing as the correction unit 106 is executed. Further, the arithmetic unit 100 is not limited to one that realizes a predetermined function through cooperation of hardware and software, and may be a hardware circuit designed exclusively for realizing a predetermined function. That is, the arithmetic unit 100 can be realized by various processors such as CPU, MPU, GPU, FPGA, DSP, and ASIC.

The storage unit 110 is a recording medium for recording various information. The storage unit 110 is implemented by, for example, RAM, ROM, flash memory, SSD (Solid State Device), hard disk, other storage devices, or an appropriate combination thereof. In addition to the control program P executed by the calculation unit 100, the storage unit 110 stores various data used in the machine tool 1 and the like. For example, the storage unit 110 stores image data D1, profile information D2, and profile image data D3.

The input unit 120 is input means such as a keyboard, mouse, and touch panel used for inputting data and operation signals. The output unit 130 is output means such as a display used to output data.

The communication unit 140 is an interface circuit (module) for enabling data communication with an external device (not shown). For example, the communication unit 140 can perform data communication with the imaging unit 13 that captures image data.

The receiving unit 101 receives a plurality of image data D1 of the tool 11 captured by the imaging unit 13 mounted inside the machine tool 1 that processes the workpiece using the tool 11 . For example, the receiving unit 101 receives image data D1 of a plurality of projection images of the tool 11 rotated and captured from different angles. Further, the receiving unit 101 associates the received plurality of image data D1 with the identification information of the tool 11 and the identification information for identifying the positional relationship between the tool 11 and the imaging unit 13 at the time of imaging, and stores them in the storage unit 110. Let As a result, point group data and profile image data D3 regarding the target tool 11 can be formed later.

FIG. 4A is an example of image data D1 received by the receiving unit 101. FIG. In the example shown in FIG. 4A, the image data D1 includes the outer shape of the tool 11 including the blade portion 111 and the shank portion 112 . The image data D1 may be acquired by one image pickup by one image pickup unit. Further, when receiving a plurality of image data captured simultaneously from a plurality of imaging units 13 that capture different imaging ranges, the reception unit 101 synthesizes the plurality of image data captured at the same time as one image data D1. good too. Furthermore, one imaging unit may take a plurality of images while moving the tool, and synthesize the images to obtain one image data D1.

The image data D1 may be an image including the entire tool 11, or may be an image including only the blade portion of the tip portion of the tool. In such a case, a plurality of images may be acquired by automatically following the contour from the tip of the tool, and the image data D1 of the tip portion of the tool may be acquired by synthesizing the images. By doing so, the imaging time can be shortened when only the image of the tip portion of the tool is sufficient.

This image data D1 is displayed on the output unit 130, and the operator uses the input unit 120 to set the boundary between the blade portion of the tool 11 automatically set and the shank portion, which is another portion. It may be possible to For example, the operator sets (A) shown in FIG. 4A as the boundary between the blade portion and the shank portion. Further, the image data D1 may be displayed on the output unit 130 so that the operator can set a range to be subjected to processing such as analysis in the image processing apparatus 10 . For example, if the analysis target range is initially set from (A) to the lower blade, the operator can change the setting from (A) to (B) as the analysis target range. be. This makes it possible to analyze the lower blade from (B). Analysis time can be shortened by analyzing only a part of the blade. Also, even if the analysis is performed in the same amount of time, the number of analysis points can be increased. In this description, the setting of the boundary between the blade portion and the shank portion has been described, but it can also be used for setting the boundary between the tool and the tool holder.

The point cloud data forming unit 102 forms point cloud data relating to the outer shape of the tool from the plurality of image data D1 received by the receiving unit 101 . Specifically, the point cloud data forming unit 102 converts a point cloud composed of a plurality of points extracted from the contour of the tool 11 included in a plurality of image data D1 obtained for one tool 11 into a point cloud. data. For example, the point cloud data forming unit 102 extracts points at predetermined intervals from the contour of the tool 11 from the image data D1 obtained for one tool 11, and synthesizes the extracted points by matching the angle components. , point cloud data. This angle component is received from the machine tool 1 together with each image data D1. Thereby, the point cloud data forming unit 102 can form point cloud data capable of representing the three-dimensional structure of the tool 11 . Further, the point cloud data forming unit 102 stores the point cloud data formed from the image data D1 as the profile information D2 together with the identification information of the tool 11 in the storage unit 110, or stores the profile information D2 of the tool 11 in the tool management information. It is stored in association with identification information. FIG. 4B is an example of point cloud data formed by the point cloud data forming unit 102 .

The profile image forming unit 103 forms profile image data D3 representing a two-dimensional or three-dimensional profile of the tool 11 from the point cloud data. Specifically, the profile image forming unit 103 can use, as the profile image data D3, a contour of a plane obtained by cutting a part of the three-dimensional structure formed by the point cloud data. Also, the profile image forming unit 103 may use the outline itself of the three-dimensional structure formed by the point cloud data of the profile information D2 as the profile image data D3. Furthermore, the profile image data D3 may be two-dimensional tool data in DXF format, which is a two-dimensional CAD data format, for example, generated from the profile information D2. Also, the profile image forming unit 103 causes the storage unit 110 to store the profile image data D3 formed from the point cloud data together with the identification information of the tool 11 . FIG. 4C is an example of profile image data D3 that three-dimensionally represents the tool formed by the profile image forming unit 103 .

The setting unit 104 sets information about the blade portion 111 of the tool 11 . Specifically, the setting unit 104 uses the point cloud data or profile image data D3 of the profile information D2 stored in the storage unit 110 to set the region of the blade portion 111 and the area of the blade portion 111 as the information of the blade portion 111. Information about size such as length and diameter can be set. Alternatively, the setting unit 104 can set the boundary between the blade portion 111 and the shank portion 112 of the tool 11 as information on the blade portion 111 . In addition, the setting unit 104 may set both information regarding the area and size of the blade portion 111 and the boundary between the blade portion 111 and the shank portion 112 as the information on the blade portion 111 .

For example, the setting unit 104 refers to data such as the shape of each type of tool 11, the shape of the blade portion 111, or the shape of the shank portion 112, which is set in advance as a reference, and sets the point cloud data or the profile image data of the profile information D2. From D3, the area and size of the blade portion 111 of the target tool 11 or the boundary between the blade portion 111 and the shank portion 112 can be set.

The memory processing unit 105 adds the information of the blade portion 111 set by the setting unit 104 to the profile information D2 as the information of the tool 11, and causes the storage unit 110 to store the information. Specifically, the memory processing unit 105 adds the information of the blade part 111 to the profile information D2 of the target tool 11 .

At this time, the correction unit 106 corrects the information on the blade unit 111 as necessary. Specifically, the point cloud data formed by the point cloud data forming unit 102 or the profile image data D3 formed by the profile image forming unit 103 are displayed on the display, which is the output unit 130 . An operator or the like can input a signal for correcting the information of the blade portion 111 with respect to the displayed data. For example, the correction unit 106 corrects the information of the blade portion 111, which is the boundary position between the blade portion 111 and the shank portion 112 of the tool 11, based on a signal input by an operator or the like via the input portion 120. FIG. Alternatively, the correction unit 106 corrects the information of the cutting edge 111 of the tool 11 based on the signal input by the operator or the like via the input unit 120 . As a result, in the image processing device 10, the information of the blade portion 111 related to the tool 11 used in the machine tool 1 is registered in the profile information D2.

<Profile information registration processing>
The process of registering the profile information D2 in the image processing apparatus 10 according to the first embodiment will be described with reference to the flowchart shown in FIG. The process of registering the profile information D2 is started when a new tool 11 is installed on the spindle 113 of the machine tool 1 and becomes ready for use. This is because the image is captured while the tool 11 is attached to the spindle 113, so the position of the tool 11 can be controlled accurately and easily. First, the receiving unit 101 receives image data D1 obtained by imaging the tool 11 attached to the spindle 113 by the imaging unit 13, and stores the image data D1 in the storage unit 110 (S01).

The point cloud data forming unit 102 uses the image data D1 of the tool 11 received in step S01 to form point cloud data composed of a plurality of point groups extracted from the outer shape of the tool 11 (S02). Further, the point cloud data forming unit 102 stores the formed point cloud data in the storage unit 110 as the profile information D2.

The profile image forming unit 103 forms profile image data D3 using the point group data of the tool 11 formed in step S02 (S03). Also, the profile image forming unit 103 causes the storage unit 110 to store the formed profile image data D3.

The setting unit 104 sets information on the cutting edge 111 of the tool 11 using the point cloud data formed in step S02 or the profile image data D3 of the tool 11 formed in step S03 (S04). Note that when setting the information of the blade portion 111 using point cloud data, the formation of the profile image data D3 in step S03 may be omitted.

When the information of the blade portion 111 set in step S04 needs to be corrected (YES in S05), the correction section 106 corrects the profile information D2 (S06). Specifically, when a worker's signal is input via the input unit 120, the correction unit 106 corrects the profile information D2 according to the input signal.

The storage processing unit 105 adds and registers the information of the blade portion 111 set in step S04 or the information of the blade portion 111 corrected in step S06 to the profile information D2 stored in the storage unit 110 (S07).

As described above, according to the image processing apparatus 10 according to the embodiment, when a new tool 11 is used in the machine tool 1, the profile image data D3 and the profile information D2 are easily formed in a short time. becomes possible. As a result, the machine tool 1 can perform an interference check using the profile image data D3 and the profile information D2, and can easily process a workpiece with high accuracy.

[Embodiment 2]
<Machine tool and image processing device>
In the machine tool 1 and the image processing apparatus 10 according to Embodiment 1, an image is acquired at the timing when the new tool 11 is stored in the machine tool 1 and the new tool 11 becomes available, Profile information D2 was registered. On the other hand, in the machine tool and the image processing apparatus according to the second embodiment, the tool 11 for which the profile information D2 has already been registered acquires a new image at a predetermined timing when wear, damage, etc. occurs due to use. and updates the profile information D2.

The machine tool according to the second embodiment has the same configuration as the machine tool 1 described above with reference to FIGS. Since it has the same configuration as the processing device 10, it will be described with reference to FIGS.

The imaging unit 13 of the machine tool 1 according to Embodiment 2 captures images of the tool at predetermined intervals while the same tool is attached to the spindle of the machine tool 1 . This makes it possible to compare image data captured at a plurality of different timings in the image processing device 10 and identify changes in the tool 11 such as wear. For example, the imaging unit 13 may periodically capture an image of the tool 11 at the timing when each tool 11 has been used a predetermined number of times. Alternatively, the imaging unit 13 may capture an image of the tool 11 when the value obtained by some sensor in the machine tool 1 is out of a predetermined range.

Here, it is preferable that the position and posture of the imaging unit 13 be fixed at the timing of each imaging. In this case, by controlling the movement of the main shaft 113 to which the tool 11 is attached, it is possible to finely adjust the focus during imaging. By fixing the imaging unit 13 and using the movement mechanism of the spindle 113 in this manner, a movement mechanism or the like is not required, and the machine tool 1 can be prevented from becoming large.

Further, the correction unit 106 of the image processing apparatus 10 according to Embodiment 2 corrects the profile information D2 determined based on the image data D1 of the tool 11 obtained by the previous imaging by the imaging unit 13 in the subsequent imaging. Correction is performed using point cloud data formed based on the obtained new image data D1 of the tool 11 . For example, the output unit 130 outputs the previous profile information D2 stored in the storage unit 110 to the output unit 130, and also outputs the profile information D2 generated from the newly obtained image data D1 side by side. This makes it easier for the operator or the like to grasp the change in the tool 11 . An operator or the like can compare these displayed data and input a signal for correcting the information of the blade portion, which is a part of the profile information D2. Alternatively, it may be possible to input a signal for correcting tool wear on the tool management screen. The correcting unit 106 corrects information about the blade portion included in the tool management information managed on the profile information D2 and the tool management screen, based on a signal newly input by the operator or the like via the input unit 120 . Also, the storage processing unit 105 registers the new profile information D2 corrected by the correction unit 106 in the storage unit 110 . As a result, the state of the tool 11 is updated based on the profile information D2. Further, in the image processing device 10, when it is detected from the updated new profile information D2 that the tool 11 is worn to the extent that replacement is required, the output unit 130 is notified that it is time to replace the tool. may be displayed.

<Profile information update processing>
A process of updating the profile information D2 in the image processing apparatus 10 according to the first embodiment will be described with reference to the flowchart shown in FIG. First, the receiving unit 101 receives new image data D1 captured by the imaging unit 13 and stores it in the storage unit 110 (S11).

The point cloud data forming unit 102 forms new point cloud data using the new image data D1 of the tool 11 received in step S11 (S12). Further, the point cloud data forming unit 102 causes the storage unit 110 to store the newly formed point cloud data as new profile information D2.

The profile image forming unit 103 forms new profile image data D3 using the new point group data of the tool 11 formed in step S12 (S13). Also, the profile image forming unit 103 causes the storage unit 110 to store the newly formed profile image data D3.

The setting unit 104 uses the new point cloud data formed in step S12 or the new profile image data D3 of the tool 11 formed in step S13 to set the current information on the cutting edge of the tool 11 (S14 ). Note that when setting the information of the blade portion 111 using the point cloud data, the formation of the new profile image data D3 in step S13 may be omitted.

The correcting unit 106 corrects the profile information D2 when it is necessary to correct the blade information set in step S14 (YES in S15) (S16). Specifically, when a signal is input from the operator via the input unit 120 for the profile information D2 or the like displayed for comparison on the output unit 130, the correction unit 106 performs Correct the profile information D2.

The storage processing unit 105 causes the storage unit 110 to store the information of the new blade portion 111 set in step S14 or the information of the new blade portion 111 corrected in step S16 (S17). Thereby, in the image processing device 10, new profile information D2 obtained regarding the current state of the tool 11 is registered.

Further, the image processing apparatus 10 determines that the current tool 11 in the state specified by the registered new profile information D2 has not reached the preset life of the tool 11 (NO in S18), and new image data (YES in S19), the processing of steps S11 to S19 is repeated. For example, in the image processing device 10, the storage unit 110 stores data for determining whether the tool 11 has reached the end of its life.

On the other hand, if it is time to end the process in the machine tool 1 (YES in S20), the process of updating the profile information D2 ends.

In addition, when the current tool 11 in the state specified by the registered new profile information D2 has reached the preset life of the tool 11 (YES in S18), the image processing apparatus 10 determines the tool replacement timing. is displayed on the output unit 130 (S21), and the process of updating the profile information D2 is terminated.

The sequence diagram shown in FIG. 7 will be used to explain the processing in the machine tool 1 when executing the processing for updating the profile information D2 described above with reference to FIG. First, the image processing device 10 transmits a position control signal for adjusting the position of the tool 11 to the NC device 17 (S31). Thereby, the NC device 17 controls the spindle 113 and adjusts the position of the tool 11 (S32).

The image processing apparatus 10 also transmits an illumination control signal S33 for dimming the illumination to the illumination unit 14 (S33). Thereby, the illumination unit 14 is dimmed (S34).

After that, the image processing device 10 transmits a shooting control signal to the imaging section 13 (S35). Thereby, the imaging unit 13 images the image data D1 of the tool 11 (S36). The imaging unit 13 also outputs the captured image data D<b>1 to the image processing device 10 .

The image processing apparatus 10 uses the acquired image data D1 to process the image and update the data as described above with reference to FIG. 6 (S38).

When continuing the processing, the image processing device 10 transmits a new position control signal to the NC device 17 (S39). The NC device 17 also controls the spindle 113 to adjust the position of the tool 11 (S40). The processing from steps S35 to S40 surrounded by a dashed line in FIG. 7 is repeated until the update processing of the profile information D2 is completed.

Alternatively, when the timing for replacing the tool 11 is reached, the image processing device 10 transmits a position control signal to the NC device 17 to control the position for replacement (S41). As a result, the NC device 17 controls the spindle 113 and adjusts the tool 11 to a position for exchanging (S42). After that, in the machine tool 1, the tool 11 is replaced.

In this manner, the image processing apparatus 10 can register new profile information D2 including changes in the tool such as wear and damage. Therefore, by using the new profile information D2, the machine tool 1 can accurately reflect deterioration and deviation such as wear and damage of the tool 11 due to long-term use, and can prevent a decrease in machining accuracy.

[Modification 1]
In the machine tool 1 described above in Embodiment 1, the imaging unit 13 has been described as capturing a transmission image of the tool 11 . Alternatively, the imaging unit 13 may capture reflected light. Specifically, the illumination unit 14 is arranged such that the light emitting surface is oriented in the same direction as the light receiving surface of the imaging unit 13 . may be configured to be incident on the As a result, the imaging unit 13 can obtain an image of the surface of the tool 11 , and can accurately grasp the damage or the like on the tool 11 .

[Modification 2]
In the machine tool 1 and the image processing apparatus 10 described above, the acquired image data D1 is used to register and update the profile information D2. On the other hand, in the image processing device 10, the tool currently held by the holding section 12 of the machine tool 1 may be identified using the acquired image data D1. In this case, the image processing apparatus 10 stores tool list data, which is information on a plurality of tools that can be used in the machine tool 1, in the storage unit 110, and the calculation unit 100 includes an identification unit for identifying the tools.

The tool list data includes, for example, tool model numbers as identification information for each tool. In addition, the tool list data includes at least the outer diameter and length of the blade portion of the tool as information on the shape of the tool for each model number. Furthermore, for a tool having a different outer diameter depending on the part of the cutting edge 111 of the tool, the information about the shape of the tool can include information about the longitudinal direction of the part of the tool that specifies the part. The identification unit determines the model number of the tool attached to the spindle 113 from the point cloud data or the profile image data D3 and the tool list data. This makes it possible to easily identify the tool held on the spindle 113 .

[Modification 3]
In the image processing apparatus 10 described above, the setting unit 104 sets information on the blade portion 111 of the tool 11 . On the other hand, the profile image forming section 103 may set the information of the blade section 111 using a learning model that has been learned in advance by machine learning. Specifically, this learning model is obtained from a set of profile image data of the tool, boundary positions of the blade portion 111 and the shank portion 112 based on actual measurement of the tool 11, and outline profile data based on actual measurement of the tool 11. The parameters obtained by learning using the teacher data are applied. This learning model receives image data of the tool 11 and outputs profile information D2 such as boundary positions of the blade portion 111 and the shank portion 112 of the tool 11 and profile image data D3. This enables the profile information D2 to be registered using the learning model.

As described above, the above embodiments have been described as examples of the technology disclosed in the present application. However, the technology in the present disclosure is not limited to this, and can also be applied to embodiments in which modifications, replacements, additions, omissions, etc. are made as appropriate.

By using the technology disclosed in this application, it becomes possible to automatically generate a 3D model of a tool for interference check. Moreover, by using the technology disclosed in the present application, it becomes possible to detect chip entanglement during tool change. Also, by using the technology disclosed in the present application, tool correction can be automated. Moreover, by using the technology disclosed in the present application, it is possible to perform quantitative tool management by detecting chipping of the rake face and measuring the amount of wear on the flank face.

The image processing device, machine tool, and image processing method described in all claims of the present disclosure are realized by cooperation with hardware resources such as processors, memories, and programs.

The image processing device, machine tool, and image processing method of the present disclosure are useful, for example, in facilitating interference checks of machine tools.

1 Machine tool 11 Tool 12 Holding unit 13 Imaging unit 14 Illuminating unit 10 Image processing device 100 Calculation unit 101 Receiving unit 102 Point cloud data forming unit 103 Profile image forming unit 104 Setting unit 105 Storage processing unit 106 Correcting unit 110 Storage unit 120 Input Unit 130 Output unit 140 Communication unit D1 Image data D2 Profile information D3 Profile image data P Control program

Claims (1)

A method for processing an image captured by a machine tool that processes a workpiece using a tool and inspecting the condition of the tool while the tool is in the machine tool, comprising:
a first step of moving the tip of the tool in front of the imaging unit while the tool is attached to the spindle, and imaging only the tip of the tool ;
a second step of capturing a plurality of images by automatically following the contour from the tip of the tool;
Analyzing the plurality of images to extract a plurality of points, and forming one point cloud data of the tool related to the profile of the one tool from the synthesis of the plurality of images captured in the second step 3 steps;
a fourth step of machining the workpiece with the tool attached to the spindle, and moving the tip of the tool in front of the imaging unit while the tool is attached to the spindle after machining;
a fifth step of automatically following the contour from the tip of the tool and capturing a plurality of images;
An inspection method for inspecting the state of a tool, comprising a sixth step of analyzing the image captured in the fifth step and comparing it with the point cloud data.

Images