A tool geometry visual measurement method

Abstract

This invention provides a method for visually measuring the geometric parameters of a cutting tool, belonging to the field of machine tool measurement technology. It includes a phase synchronization triggering step, a multispectral time-division imaging step, a spatial coordinate calibration step, a multispectral fusion and edge extraction step, and a geometric calculation and compensation step. In the multispectral time-division imaging step, in response to a synchronization triggering command, the visible light channel and near-infrared channel in the multispectral illumination unit alternately flash in a time-division manner, and the original images of the cutting tool in different spectral channels are acquired through a narrow-band filter at the front end of a binocular industrial camera. In the multispectral fusion and edge extraction step, the acquired original images of the cutting tool in multiple spectral channels are weighted and fused pixel-by-pixel. During fusion, the tool substrate contour information presented by the near-infrared channel image and the cutting edge coating texture features presented by the visible light channel image are used to identify and remove cutting fluid droplets and reflective artifacts attached to the cutting edge, extracting the true geometric edge contour of the cutting tool.

Description

Technical Field

[0001] This invention relates to the field of measurement or monitoring devices for machine tools, and in particular to a method for visually measuring the geometric parameters of cutting tools. Background Technology

[0002] As modern manufacturing moves towards higher precision and intelligence, CNC machine tools are increasingly widely used in aerospace, precision molds, and semiconductor equipment. During high-speed cutting, the geometric parameters of the cutting tool, such as diameter, length, and radial runout, directly determine the machining accuracy of the parts. To ensure machining quality, precise measurement of the cutting tool's condition is essential.

[0003] Current tool measurement technologies are mainly divided into two categories: external measurement and internal measurement. Traditional external measurement typically uses offline tool setters, which can provide high static measurement accuracy. However, since the parameters are measured when the tool is stationary, they cannot truly reflect the dynamic characteristics of the tool under the high-speed rotation of the CNC spindle. In actual machining, due to the influence of centrifugal force, spindle temperature rise and thermal expansion, and radial runout generated by high-speed rotation, the dynamic cutting diameter of the tool deviates significantly from the static value, resulting in out-of-tolerance part dimensions.

[0004] To address this issue, the industry has begun experimenting with integrating visual measurement systems within machine tools. However, existing in-machine vision solutions still face the following significant technical challenges in practical engineering applications: First, the image acquisition and spindle rotation were not precisely synchronized, resulting in severe motion blur in the image during high-speed rotation, making it impossible to lock the geometric features of the cutting edge at specific angles. Secondly, the cutting zone of a machine tool is filled with cutting fluid droplets, oil mist, metal chips, and sparks generated during cutting. Droplets adhering to the cutting edge can alter the refraction path of light, producing bright reflective artifacts. These artifacts are often identified as false edges in conventional vision algorithms, leading to errors in measurement results.

[0005] Finally, due to the thermal deformation of structural components caused by long-term operation of machine tools, the spatial pose of the measuring device relative to the tool will drift. Furthermore, traditional measurement schemes lack real-time feedback from the machine tool's CNC system, making it impossible to achieve closed-loop control from measurement to automatic compensation. Summary of the Invention

[0006] The present invention aims to solve the above-mentioned technical problems by providing a method for visually measuring tool geometry parameters.

[0007] The technical solution of this invention is a method for visually measuring tool geometry parameters, applied in an environment including a synchronous trigger control unit, a multispectral illumination unit, a binocular industrial camera, and a machine tool CNC system, comprising: Phase synchronization triggering steps: The zero-position signal and angular displacement pulse of the spindle encoder are acquired through the synchronization triggering control unit to establish the correspondence between the spindle rotation phase and the camera triggering time; when the spindle carrying the tool rotates to the preset measurement phase angle, the synchronization triggering control unit sends a synchronization triggering command to the multispectral illumination unit and the binocular industrial camera to obtain the tool image; Multispectral time-division imaging step: In response to the synchronous trigger command, the visible light channel and near-infrared channel in the multispectral illumination unit are alternately flashed in a time-division manner, and the original images of the tool in different spectral channels are acquired through the narrow band filter at the front end of the binocular industrial camera; Spatial coordinate calibration steps: Set a standard ball at a preset position on the machine tool worktable. Before measurement, drive the binocular industrial camera through the linear axis of the machine tool to perform multi-view scanning on the standard ball, establish the affine transformation matrix between the machine tool mechanical coordinate system and the pixel coordinate system of the binocular industrial camera, and use the affine transformation matrix to correct the spatial pose deviation. Multispectral fusion and edge extraction steps: The original tool images from multiple spectral channels are weighted and fused pixel by pixel. During fusion, the tool substrate contour information presented by the near-infrared channel image and the edge coating texture features presented by the visible light channel image are used to identify and remove cutting fluid droplets and reflective artifacts attached to the tool tip edge, and extract the true geometric edge contour of the tool. Geometric calculation and compensation steps: Based on the actual geometric edge contour line, calculate at least one geometric parameter among the tool radial runout, actual cutting diameter, and edge wear, and feed the geometric parameter back to the machine tool CNC system, which then corrects the tool compensation value in the machining program.

[0008] In one implementation, there are multiple preset measurement phase angles, and the angular intervals between adjacent measurement phase angles are equal.

[0009] In one implementation, in the phase synchronization triggering step, the synchronization triggering control unit calculates the real-time rotational speed of the current spindle based on the received angular displacement pulse frequency, and calculates the trigger delay time based on the preset measured phase angle, so as to compensate for the hysteresis error caused by signal transmission and hardware response.

[0010] In one embodiment, the multispectral illumination unit includes a plurality of LED emitters arranged in a ring array, with emitters of different spectral channels arranged alternately, and the binocular industrial camera is located on the central axis of the ring array.

[0011] In one embodiment, in the multispectral time-division imaging step, the center wavelength range of the near-infrared channel is 780nm to 950nm.

[0012] In one implementation, during the multispectral time-division imaging step, the visible light channel extracts the coating boundary line on the tool surface through high-contrast illumination.

[0013] As one implementation method, the spatial coordinate calibration step includes: driving the linear axis of the machine tool to make the standard sphere perform multi-point displacement within the overlapping field of view of the binocular industrial camera, obtaining the image coordinates of the sphere center under different machine coordinates, and using the least squares method to fit and obtain the affine transformation matrix.

[0014] As one implementation method, in the multispectral fusion and edge extraction step, the image is fused in the following way: a weight coefficient is set according to the gray-level gradient value of the pixel. For regions with drastic gradient changes and conforming to the morphological characteristics of droplets, the weight ratio of the near-infrared channel is increased; for high-frequency texture regions, the weight ratio of the visible light channel is increased.

[0015] As one implementation method, in the geometric calculation and compensation step, the method for calculating the actual cutting diameter is as follows: extract the coordinates of edge points under multiple phase angles within one revolution of the tool, construct a dynamic envelope circle, and determine the diameter of the envelope circle as the actual cutting diameter.

[0016] As one implementation, the process of feeding the geometric parameters back to the machine tool CNC system includes: transmitting the calculated geometric parameters to the external data register of the machine tool CNC system, triggering the machine tool to automatically update the tool compensation value.

[0017] The advantages of this invention compared to existing technologies are that, by synchronously triggering the phase-locking between the control unit and the spindle encoder, it can accurately acquire the dynamic deformation and centrifugal runout of the tool at the actual machining speed, resulting in measurement results that more closely resemble the actual machining state. Utilizing multispectral time-division imaging and weighted fusion algorithms, it leverages the penetrating characteristics of near-infrared light through cutting oil mist, effectively eliminating artifacts from droplets adhering to the cutting edge and interference from metallic reflections, ensuring contour extraction accuracy even in harsh machining environments. Through in-situ calibration of a standard sphere and an affine transformation matrix, a mapping between the machine tool coordinate system and the camera pixel coordinate system is established, enabling real-time compensation for spatial pose deviations caused by thermal deformation during long-term machine tool operation or minute displacements of the camera support. Measurement data is directly fed back to the CNC system and automatically corrects tool compensation values ​​without machine downtime, significantly improving production efficiency while enhancing the dimensional consistency of large and complex parts. Attached Figure Description

[0018] Figure 1 A flowchart illustrating the tool geometry parameter visualization measurement method provided for embodiments of the present invention. Detailed Implementation

[0019] The above and other embodiments and advantages of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0020] In one implementation, such as Figure 1 As shown.

[0021] The tool geometry parameter visualization measurement method provided in this embodiment is applied in an environment including a synchronous trigger control unit, a multispectral illumination unit, a binocular industrial camera, and a machine tool CNC system. It includes: S1 Phase Synchronous Triggering Step: The synchronous trigger control unit acquires the zero-position signal and angular displacement pulse from the spindle encoder, establishing a correspondence between the spindle rotation phase and the triggering time of the binocular industrial camera; when the spindle carrying the tool rotates to a preset measurement phase angle, the synchronous trigger control unit sends a synchronous trigger command to the multispectral illumination unit and the binocular industrial camera to obtain a tool image; S2 Multispectral Time-Division Imaging Step: In response to the synchronous trigger command, the visible light channel and near-infrared channel in the multispectral illumination unit alternately flash in a time-division manner, and the original tool images of different spectral channels are acquired through a narrowband filter at the front end of the binocular industrial camera; S3 Spatial Coordinate Calibration Step: A standard is set at a preset position on the machine tool worktable. Before measurement, the standard sphere is scanned from multiple angles by a binocular industrial camera driven by the linear axis of the machine tool. An affine transformation matrix is ​​established between the machine tool's mechanical coordinate system and the pixel coordinate system of the binocular industrial camera. The affine transformation matrix is ​​used to correct spatial pose deviation. S4 Multispectral Fusion and Edge Extraction Step: The original tool images from multiple spectral channels are weighted and fused pixel by pixel. During fusion, the tool substrate contour information presented by the near-infrared channel image and the edge coating texture features presented by the visible light channel image are used to identify and remove cutting fluid droplets and reflective artifacts attached to the tool tip edge, and extract the true geometric edge contour line of the tool. S5 Geometric Calculation and Compensation Step: Based on the true geometric edge contour line, at least one geometric parameter among the tool radial runout, actual cutting diameter, and edge wear is calculated. The geometric parameter is fed back to the machine tool CNC system, which corrects the tool compensation value in the machining program.

[0022] In this embodiment, the synchronous trigger control unit can use a high-performance microcontroller as the core processor. Its input terminal is connected to the signal output interface of the machine tool spindle encoder to capture the zero-position signal and angular displacement pulse of the spindle rotation in real time. The multispectral illumination unit consists of a visible light LED array and a near-infrared LED array arranged in a ring, including visible light channels and near-infrared channels. It is installed inside the machine tool protective cover and points towards the tool cutting area. The binocular industrial camera is symmetrically arranged below the spindle, also installed inside the machine tool protective cover and pointing towards the tool cutting area, so that it forms a relative displacement with the tool moving with the spindle. The front end of the binocular industrial camera is equipped with a narrowband filter of a specific wavelength, allowing only the selected spectrum to pass through, in order to suppress interference from ambient stray light and cutting sparks. Its trigger port is connected to the output terminal of the synchronous trigger control unit through a shielded cable. The machine tool CNC system acts as the data interaction center, receiving measurement results and performing corrections via Ethernet or fieldbus. This arrangement differs from the traditional external tool setter; it directly utilizes the machine tool's own structural components as the measurement reference, solving the problem that the thermal expansion and dynamic deformation of the tool caused by rotation cannot be fed back in real time.

[0023] The following example demonstrates high-speed dynamic measurement of a φ20mm four-flute end mill: First, spatial coordinate calibration is performed. A standard sphere is fixedly connected at a preset position on the edge of the machine tool table, within a safe area that does not interfere with machining. Before measurement begins, the standard sphere mounted on the table is driven into the fixed field of view of the binocular industrial camera via the machine tool's linear axis, and multi-point displacement scanning is performed, resulting in multiple images at different X / Y / Z coordinates. Utilizing the principle of sphere center projection geometry, an affine transformation matrix is ​​calculated and established between the machine tool's mechanical coordinate system and the pixel coordinate system. This matrix includes rotation, translation, and scaling factors, effectively compensating for spatial pose deviations caused by thermal displacement during long-term machine tool operation or minor deformation of the binocular industrial camera bracket. During the measurement cycle, the spindle carries the tool and rotates at the actual machining speed. The synchronous trigger control unit analyzes the encoder pulses in real time, calculating the correspondence between the spindle rotation phase and the trigger moment of the binocular industrial camera. Specifically, the system determines the rotation zero position based on the encoder's zero-position signal and calculates the corresponding rotation angle based on the pulse frequency. When the system detects that a cutting edge is about to rotate to a preset measurement phase angle, such as a phase orthogonal to the optical axis of a binocular industrial camera, the synchronous trigger control unit sends pulses to the binocular industrial camera and the multispectral illumination unit. Specifically, the synchronous trigger control unit can trigger different spectral channels at the same phase angle during two consecutive cycles of spindle rotation to ensure image consistency in spatial coordinates. At this time, the multispectral illumination unit performs alternating strobe, that is, it illuminates the visible light source and the near-infrared light source successively within a short time interval, and the binocular industrial camera simultaneously exposes, acquiring two sets of images with different characteristics. In the multispectral fusion and edge extraction stage, the system performs pixel-by-pixel weighted fusion through an algorithm. Because the cutting fluid droplets generated during the cutting process have a strong refractive and reflective effect on visible light, they will form bright reflective artifacts in ordinary images, obscuring the true cutting edge. However, the near-infrared channel possesses stronger penetrating power, capable of filtering out most of the interference caused by droplet dispersion and clearly presenting the tool substrate contour information, i.e., the solid boundary of the tool. Utilizing the difference between the high transmittance of the near-infrared band at the liquid interface and its high reflectivity on the metal surface, the algorithm determines whether a point is a droplet artifact by comparing the grayscale gradient difference between the two images at the same coordinate. Meanwhile, the visible light channel image preserves the texture features of the cutting edge coating caused by coating peeling or chipping. The algorithm automatically eliminates noise points caused by droplets by comparing the pixel gradient changes of the two channels, ultimately outlining the true geometric edge contour line unaffected by oil mist. Finally, based on the extracted contour line, the system calculates the radial runout by comparing the trajectory differences between multiple cutting edges and measures the actual cutting diameter of the tool under centrifugal force and the amount of cutting edge wear compared to the initial state. The calculation results are directly fed back to the machine tool CNC system, which automatically adjusts the D and H values ​​in the tool compensation table.

[0024] The above closed-loop process enables real-time detection of tool status without stopping or changing tools, greatly improving the dimensional consistency of large and complex parts during high-speed cutting.

[0025] In one embodiment, the tool geometry parameter visualization measurement method has multiple preset measurement phase angles, and the angle interval between adjacent measurement phase angles is equal.

[0026] In this embodiment, multiple preset measurement phase angles can be set during the measurement process. For example, for a four-flute end mill, 8 or 16 equally spaced sampling points can be preset. Through this equally spaced sampling, the system can acquire the projected profiles of different positions within one revolution of the tool, thereby more accurately analyzing the roundness deviation of the tool and the consistency of each cutting edge, avoiding the limitation that a single phase measurement cannot reflect the overall appearance of the tool.

[0027] In one embodiment, in the phase synchronization triggering step of the tool geometry parameter visualization measurement method, the synchronization triggering control unit calculates the real-time rotational speed of the current spindle based on the received angular displacement pulse frequency, and calculates the trigger delay time based on the preset measurement phase angle to compensate for the hysteresis error caused by signal transmission and hardware response.

[0028] In this embodiment, considering the hardware delay between the signal generated by the spindle encoder and the actual response from the light source and camera, the synchronous trigger control unit calculates the current real-time spindle speed based on the frequency of the angular displacement pulses collected in real time. The system automatically calculates a trigger delay compensation time to ensure that at high speeds, the tool phase is exactly at a preset angle at the instant the camera shutter opens, eliminating angular offset errors in dynamic measurements.

[0029] In one embodiment, the multispectral illumination unit of the tool geometry parameter visualization measurement method includes several LED emitters distributed in a ring array, with emitters of different spectral channels arranged alternately, and the binocular industrial camera is located on the central axis of the ring array.

[0030] In this embodiment, the multispectral illumination unit is structured as a ring array of several visible light LEDs and near-infrared LEDs arranged alternately. This design ensures that light can uniformly illuminate the cutting edge of the tool from multiple angles, reducing shading. A binocular industrial camera is located on the central axis of this ring array, achieving a near-coaxial illumination effect and facilitating the acquisition of high-contrast images of the cutting edge.

[0031] In one embodiment, in the multispectral time-division imaging step of the tool geometry parameter visualization measurement method, the center wavelength range of the near-infrared channel is 780nm to 950nm.

[0032] In this embodiment, the near-infrared channel uses a filter and light source with a center wavelength of 780nm to 950nm. Infrared light in this band has good penetration performance for emulsions and oil mist, while effectively avoiding interference from visible light illumination and cutting sparks in the workshop, ensuring that a clear backlight outline of the tool substrate can still be captured even in extremely harsh machine environments.

[0033] In one embodiment, in the multispectral time-division imaging step of the tool geometry parameter visualization measurement method, the visible light channel extracts the coating boundary line on the tool surface through high-contrast illumination.

[0034] In this embodiment, the visible light channel is primarily used to extract the fine texture of the tool surface. Utilizing high-brightness visible light, the system can clearly identify the grayscale differences between the tool coating and the substrate, as well as the areas of coating wear. By analyzing these features, the system can automatically locate micro-chipping points on the cutting edge and quantify the area of ​​coating detachment.

[0035] In one embodiment, the spatial coordinate calibration step of the tool geometry parameter visualization measurement method includes: driving the linear axis of the machine tool to make the standard sphere perform multi-point displacement within the overlapping field of view of the binocular industrial camera, obtaining the sphere center image coordinates under different machine coordinates, and using the least squares method to fit and obtain the affine transformation matrix.

[0036] In this embodiment, during spatial coordinate calibration, the CNC-controlled linear axis of the machine tool moves a standard sphere along a matrix trajectory within the overlapping field of view of the binocular camera. The system records the machine tool coordinates and the sphere's center pixel coordinates in the image corresponding to each position. Affine transformation matrix between the two is calculated using the least squares method for fitting. This matrix achieves a mapping from pixel size to physical size and compensates for distortion caused by the camera's mounting tilt.

[0037] In one embodiment, in the multispectral fusion and edge extraction step of the tool geometry parameter visualization measurement method, the image is fused in the following way: a weight coefficient is set according to the gray-level gradient value of the pixel. For regions with drastic gradient changes and conforming to droplet morphological characteristics, the weight ratio of the near-infrared channel is increased; for high-frequency texture regions, the weight ratio of the visible light channel is increased.

[0038] In this embodiment, during the multispectral fusion step, the algorithm performs regional attribute analysis on the image. If a region is detected to have a circular grayscale gradient that changes rapidly, consistent with droplet morphology, the weight of the visible light channel in that region is reduced, while the weight of the near-infrared channel is increased. Conversely, in high-frequency regions where wear texture needs to be identified, the weight of the visible light channel is increased. Through this adaptive weighting, a clean and information-rich fused contour map is obtained.

[0039] In one embodiment, in the geometric calculation and compensation step of the tool geometry parameter visualization measurement method, the method for calculating the actual cutting diameter is as follows: extract the coordinates of edge points under multiple phase angles within one revolution of the tool, construct a dynamic envelope circle, and determine the diameter of the envelope circle as the actual cutting diameter.

[0040] In this embodiment, when measuring the actual cutting diameter, due to the centrifugal force and runout of the tool under high-speed rotation, the system extracts the edge coordinates of all sampled phase points within one revolution of the tool. The algorithm projects these coordinate points onto the same coordinate system and constructs a minimum circumcircle, i.e., the envelope circle, through a dynamic envelope algorithm. The diameter of this envelope circle represents the dynamic diameter of the tool when it actually participates in cutting, and compared with static tool setting, it more accurately reflects the machining dimensions.

[0041] In one embodiment, the process of feeding back the geometric parameters of the tool geometry parameter visualization measurement method to the machine tool CNC system includes: transmitting the calculated geometric parameters to the external data register of the machine tool CNC system, triggering the machine tool to automatically update the tool compensation value.

[0042] In this embodiment, after the measurement results are generated, the system writes the data to the external data register of the machine tool's CNC system via Ethernet. At this time, the system triggers an automatic update command, and the machine tool automatically rewrites the D and H values ​​in the tool compensation table according to the measured actual cutting diameter. That is, radius / diameter compensation and length compensation. This method does not require manual intervention and directly realizes automatic error compensation during the machining process.

[0043] The specific embodiments described above further illustrate the inventive purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. In particular, it should be noted that any modifications, equivalent substitutions, or improvements made by those skilled in the art within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for visually measuring tool geometry parameters, characterized in that, Applications include environments containing synchronous trigger control units, multispectral illumination units, binocular industrial cameras, and machine tool CNC systems, including: Phase synchronization triggering steps: The zero-position signal and angular displacement pulse of the spindle encoder are acquired through the synchronization triggering control unit to establish the correspondence between the spindle rotation phase and the camera triggering time; when the spindle carrying the tool rotates to the preset measurement phase angle, the synchronization triggering control unit sends a synchronization triggering command to the multispectral illumination unit and the binocular industrial camera to obtain the tool image; Multispectral time-division imaging step: In response to the synchronous trigger command, the visible light channel and near-infrared channel in the multispectral illumination unit are alternately flashed in a time-division manner, and the original images of the tool in different spectral channels are acquired through the narrow band filter at the front end of the binocular industrial camera; Spatial coordinate calibration steps: Set a standard ball at a preset position on the machine tool worktable. Before measurement, drive the binocular industrial camera through the linear axis of the machine tool to perform multi-view scanning on the standard ball, establish the affine transformation matrix between the machine tool mechanical coordinate system and the pixel coordinate system of the binocular industrial camera, and use the affine transformation matrix to correct the spatial pose deviation. Multispectral fusion and edge extraction steps: The original tool images from multiple spectral channels are weighted and fused pixel by pixel. During fusion, the tool substrate contour information presented by the near-infrared channel image and the edge coating texture features presented by the visible light channel image are used to identify and remove cutting fluid droplets and reflective artifacts attached to the tool tip edge, and extract the true geometric edge contour of the tool. Geometric calculation and compensation steps: Based on the actual geometric edge contour line, calculate at least one geometric parameter among the tool radial runout, actual cutting diameter, and edge wear, and feed the geometric parameter back to the machine tool CNC system, which then corrects the tool compensation value in the machining program.

2. The tool geometry parameter visualization measurement method according to claim 1, characterized in that, The preset measurement phase angles are multiple, and the angular intervals between adjacent measurement phase angles are equal.

3. The method for visually measuring tool geometry parameters according to claim 1, characterized in that, In the phase synchronization triggering step, the synchronization triggering control unit calculates the real-time rotational speed of the current spindle based on the received angular displacement pulse frequency, and calculates the trigger delay time based on the preset measured phase angle to compensate for the hysteresis error caused by signal transmission and hardware response.

4. The tool geometry parameter visualization measurement method according to claim 1, characterized in that, The multispectral illumination unit includes several LED emitters arranged in a ring array, with emitters of different spectral channels arranged alternately, and the binocular industrial camera is located on the central axis of the ring array.

5. The method for visually measuring tool geometry parameters according to claim 1, characterized in that, In the multispectral time-division imaging step, the center wavelength range of the near-infrared channel is 780nm to 950nm.

6. The method for visually measuring tool geometry parameters according to claim 1, characterized in that, In the multispectral time-division imaging step, the visible light channel extracts the coating boundary line on the tool surface through high-contrast illumination.

7. The method for visually measuring tool geometry parameters according to claim 1, characterized in that, The spatial coordinate calibration steps include: driving the linear axis of the machine tool to make the standard sphere perform multi-point displacement within the overlapping field of view of the binocular industrial camera, obtaining the image coordinates of the sphere center under different mechanical coordinates, and using the least squares method to fit and obtain the affine transformation matrix.

8. The method for visually measuring tool geometry parameters according to claim 1, characterized in that, In the multispectral fusion and edge extraction steps, the image is fused in the following way: weight coefficients are set according to the gray-level gradient values ​​of pixels. For regions with drastic gradient changes and that conform to the morphological characteristics of droplets, the weight ratio of the near-infrared channel is increased; for high-frequency texture regions, the weight ratio of the visible light channel is increased.

9. The method for visually measuring tool geometry parameters according to claim 1, characterized in that, In the geometric calculation and compensation step, the method for calculating the actual cutting diameter is as follows: extract the coordinates of edge points under multiple phase angles within one revolution of the tool, construct a dynamic envelope circle, and determine the diameter of the envelope circle as the actual cutting diameter.

10. The method for visually measuring tool geometry parameters according to claim 1, characterized in that, The process of feeding back the geometric parameters to the machine tool CNC system includes: transmitting the calculated geometric parameters to the external data register of the machine tool CNC system, triggering the machine tool to automatically update the tool compensation value.

Images