CN116000701A - Tool detection device and method that can be integrated on tool magazine - Google Patents

Abstract

The invention discloses a tool detection device and method that can be integrated on a tool magazine. The device includes an image acquisition module, a cleaning module, a control module, etc.; the image acquisition module includes an image acquisition support device, a bottom edge image acquisition device, a side Edge image acquisition device, shell structure, etc., the lighting module is composed of two ring-shaped light sources, the control module is composed of a controller, a memory, etc.; the image acquisition support device can realize the bottom edge image acquisition device and the side edge image acquisition device to move up and down, And the side edge image acquisition device rotates throughout the circle, and then through the visual detection method, the detection of the type of tool on the tool magazine and the degree of tool wear is realized.

Description

Tool detection device and method that can be integrated on tool magazine

technical field

The invention belongs to the field of mechanical automation, in particular to a tool detection device and method which can be integrated on a tool magazine.

Background technique

High-precision, high-efficiency, multi-functional compound machining center is a main development direction of CNC equipment. The tool magazine and automatic tool changer are matched in large and medium-sized compound machining centers, which help to realize the high efficiency, high precision and high integration of the machining center. In the machining process, the wear of the tool is one of the key issues affecting the machining accuracy, machining quality and efficiency of the product. With the development of society and science and technology, the requirements for the manufacturing quality of products are getting higher and higher, which makes the detection of tool wear in the processing process more important. Therefore, during the tool change process of the tool magazine, the system can automatically judge the wear degree of the tool and replace the tool in time, which is very important for avoiding the decline in processing quality or other losses caused by excessive tool wear, while improving production efficiency and reducing labor costs. meaning.

At present, there is insufficient research on tool wear detection in China, especially the detection of tool wear in the tool magazine.

CN201822210994.6 discloses a machine tool monitoring and data measurement system, which uses strain gauges to detect the wear of the tool length in the machine tool head, but the accuracy is not enough. Most of the current detection devices and methods are aimed at the detection of machine tool tool wear, without combining the tool magazine with tool wear detection, and cannot detect tool wear and replace the tool in time while improving the accuracy of tool change in the tool magazine.

Contents of the invention

The object of the present invention is to provide a tool detection device and method that can be integrated on a tool magazine, so as to improve the accuracy of tool changing and detect tool wear in the tool magazine.

The technical solution that realizes the object of the present invention is:

A tool detection device that can be integrated on a tool magazine, including an image acquisition module and a control module; the image acquisition module includes:

The bottom edge image acquisition device is used to collect the image of the bottom edge of the tool;

The side edge image acquisition device is used to acquire the image of the tool edge measurement;

The image acquisition support device can drive the rotation and lifting of the bottom edge image acquisition device and the side edge image acquisition device at the same time; through the lifting, the bottom edge of the tool is located in the depth of field of the lens of the bottom edge image acquisition device, and the side edge image acquisition device obtains the tool The standard position picture of each edge; through rotation, the side edge image acquisition device can obtain the standard position picture of the bottom edge of the tool;

The horizontal distance of the side edge image acquisition device relative to the image acquisition support device is adjustable, and is used to adjust the tool side edge to be located within the depth of field of the side edge camera lens;

The control module is used to control the rotation and lifting of the image acquisition support device.

A tool wear detection method of a tool detection device that can be integrated on a tool magazine, including two processes of parameter acquisition and wear measurement;

The parameter acquisition process is used to enter the information of the new tool before the processing activity starts, including the model of the tool, the number of edges, the diameter and the image template;

The wear measurement process is used to return the tool to the tool magazine after processing the workpiece and take out the tool through the mechanical arm for wear detection before the next processing activity. Mask the worn tool threshold image and the unworn image template to perform The extraction of the wear area calculates the average pixel value of the wear area and converts it into the actual average wear size.

Compared with the prior art, the present invention has the remarkable advantages of:

(1) The tool detection device of the present invention has a high degree of integration, can realize fast, efficient and accurate automatic measurement of tool wear, and can be applied to different types of tool magazines.

(2) Combined with the tool magazine, it can not only improve the accuracy of the tool change in the tool magazine, but also replace the excessively worn tools in time by monitoring the wear accuracy, so as to improve the machining accuracy.

Description of drawings

Figure 1 is a diagram of the overall structure of the tool magazine and wear detection device.

Figure 2 is a diagram of the wear detection device.

Fig. 3 is a cross-sectional view of the image acquisition support device.

Figure 4 is a diagram of the image acquisition support device.

Fig. 5 is a diagram of the bottom edge image acquisition device.

Fig. 6 is a diagram of a side edge image acquisition device.

Figure 7 is a schematic diagram of the installation of the parameter calibration board. Attachment (a) is a schematic diagram of side edge camera calibration, and Figure (b) is a schematic diagram of bottom edge camera calibration.

Figure 8 is a coordinate diagram of the lens and tool.

Figure 9 is a tool positioning diagram.

Fig. 10 is a schematic diagram of tool angle and relative standard position. Accompanying drawing (a) is a drawing of the bottom edge of the tool before rotation, and drawing (b) is a drawing of the bottom edge of the tool in the standard position after rotation.

Figure 11 is a diagram of the extraction process of the bottom edge template. Accompanying drawing (a) is a drawing of the bottom edge of an unworn tool, and accompanying drawing (b) is a partial cutaway view of a cutting tool bottom edge.

Fig. 12 is a diagram of the extraction process of the side edge template. Wherein, accompanying drawing (a) is a side edge view of an unworn tool, and accompanying drawing (b) is a cutaway view of a side edge of a tool.

Figure 13 is a diagram of the process of extracting the worn part of the side edge. Attached drawing (a) is a side edge wear threshold map, drawing (b) is a side edge wear template map, drawing (c) is a side edge mask map, and drawing (d) is a side edge wear zone map.

Figure 14 is a diagram of the extraction process of the worn part of the bottom edge. Attached drawing (a) is a wear threshold diagram of the bottom edge, drawing (b) is a wear template map of the bottom edge, drawing (c) is a mask map of the bottom edge, and drawing (d) is a wear zone map of the bottom edge.

Fig. 15 is a graph of the amount of wear. The accompanying drawing (a) is a schematic diagram of the wear zone VB of the side edge of the tool, and the accompanying drawing (b) is a schematic diagram of the wear zone of the bottom edge of the tool.

Detailed ways

The present invention will be further introduced below in conjunction with the accompanying drawings and specific embodiments.

In conjunction with Fig. 1, the present invention first introduces the knives in the tool magazine of the detection object. The tool magazine includes a chain tool magazine 1, a tool magazine base 2, a tool changing arm 3, a tool handle and a tool 4; Inside the shank, the tool handle is stored in the tool magazine. When it needs to be called, the tool magazine transports the corresponding tool handle to the tool reversing position and then reverses the tool. The tool holder of the machine tool is exchanged with the tool holder of the tool magazine through the tool changing arm.

In conjunction with Fig. 2-Fig. 6, a kind of tool wear detection device 5 applied to the tool magazine of the present invention includes an image acquisition module, a cleaning module, and a control module; the image acquisition module includes an image acquisition support device 6, a bottom edge Image acquisition device 7, side edge image acquisition device 8, shell structure 9;

The image acquisition support device 6 includes a fixed frame 6-1, a vertical drive electric cylinder 6-2, a guide device 6-3, a support platform 6-4, a rotary drive device 6-5, a small gear 6-6, and a large gear 6 -7, hollow shaft sleeve 6-8, connecting shaft sleeve 6-9, encoder 6-10, encoder support shaft 6-11, encoder connecting shaft sleeve 6-12;

The vertical drive electric cylinder 6-2 is fixedly connected to the fixed frame 6-1, and the driving end of the vertical drive electric cylinder 6-2 is fixedly connected to the support platform 6-4, so that the drive support platform 6-4 can be fixed relative to the fixed frame 6-1 moves up and down, the fixed end of the guiding device 6-3 is fixedly connected to the fixed frame 6-1, and the moving end of the guiding device is fixedly connected to the support platform 6-4;

The rotary driving device 6-5 (motor) is fixedly connected under the support platform 6-4, and its drive end is fixedly connected with the pinion 6-6, and the pinion 6-6 and the bull gear 6-7 form a Gear transmission, the large gear 6-7 is rotatably connected to the hollow bushing 6-8, the hollow bushing 6-8 is fixed above the supporting platform 6-4, and the encoder supporting shaft 6-11 It is fixed under the support platform 6-4, and the hollow sleeve 6-8 is coaxial with the encoder support shaft 6-11; the encoder 6-10 is fixed on the encoder support shaft 6 -11 on;

The connecting shaft sleeve 6-9 is fixedly connected to the bull gear 6-7, the encoder connecting shaft sleeve 6-12 is fixedly connected to the lower part of the connecting shaft sleeve 6-9, and the encoder connecting shaft sleeve 6-12 Embedded in the inner cavity of the hollow shaft sleeve 6-8, the output shaft of the encoder 6-10 is fixedly connected to the encoder connecting shaft sleeve 6-12;

The axis of the bull gear 6-7, the axis of the connecting shaft sleeve 6-9, the encoder connecting shaft 6-12, and the output shaft of the encoder 6-10 are coaxial;

The bottom edge image acquisition device 7 includes a bottom edge ring light source 7-1, a bottom edge light source support part 7-2, a bottom edge camera 7-3, and a bottom edge camera support 7-4; the bottom edge camera support 7 -4 is affixed to the lower part of the connecting shaft sleeve 6-9, the bottom edge camera 7-3 is affixed to the bottom edge camera support 7-4, and the bottom edge light source support part 7-2 is affixed to the connecting shaft sleeve 6- 9, the bottom edge ring light source 7-1 is fixedly connected to the bottom edge light source supporting part 7-2 to provide light source for the bottom edge camera 7-3;

The side edge image acquisition device 8 includes a side edge ring light source 8-1, a side edge light source support 8-2, a side edge camera 8-3, and a side edge camera support; the side edge camera support includes a guide rail mounting plate 8-4. Guide rail pair 8-5 and support rod 8-6 containing the clamp; the side edge ring light source 8-1 is fixedly connected to the side edge camera 8-3 through the side edge light source support part 8-2, and is a side edge light source. The blade camera 8-3 provides the light source; the guide rail mounting plate 8-4 is affixed to the connection bushing 6-9, the guide rail pair 8-5 containing the clamp is affixed to the guide rail mounting plate 8-4, and the support rod 8 -6 is affixed to the slider of the guide rail pair 8-5 containing the clamp; the side edge camera 8-3 is affixed to the support rod 8-6.

The camera module of the image acquisition module includes a bottom edge camera 7-3 and a side edge camera 8-3, one is placed directly below the tool for collecting clear pictures of the bottom edge of the tool, and the other is placed on the side of the tool for collecting The image of the measured edge; the actual size of the tool is calculated by the pixels of the obtained image and the magnification.

The lighting module of the image acquisition module includes a bottom edge ring light source 7-1 and a side edge ring light source 8-1, which are used to provide good lighting and improve the clarity of the captured image; the ring light source is coaxial with the camera and can move up and down Obtain the best lighting; the ring light source is perpendicular to the tool, and through the light reflection on the tool surface, it is imaged inside the camera to obtain an image.

The motion support module of the image acquisition module includes a rotary drive device 6-5, a vertical drive electric cylinder 6-2, a clamp rail pair 8-5 and several components. The support module can realize the automatic adjustment of the bottom edge camera structure: up and down movement and rotation and manual adjustment of the side edge camera structure. The vertical drive electric cylinder 6-2 is used to realize the up and down movement of the bottom edge camera structure; through the up and down movement of the piston rod, the guide device 6-3 is used to push the image acquisition structure to move vertically, so that the bottom edge of the tool is positioned at the depth of field of the lens Inside, the focus of the bottom edge camera is realized; the rotary driving device 6-5 provides the rotation force for the bottom edge camera 7-3, so that the bottom edge camera structure rotates at a low speed; the rotation of the bottom edge camera 7-3 obtains the standard position of the bottom edge of the tool The picture provides the basis for subsequent image processing; the standard position is that the straight line formed by the line between the tool tip and the center of the bottom surface of the tool is parallel to the vertical edge of the bottom edge image; the side edge camera 8-3 is supported by the side edge camera The parts are fixed to the bottom edge camera structure, so that the bottom edge camera structure can drive the side edge camera structure to rotate, and the bottom edge camera 7-3 is rotated to obtain a standard position picture of each edge of the tool. At the same time, the distance between the side edge camera structure and the bottom edge camera structure can be manually adjusted through the guide rail; after reaching a suitable position (the side edge of the tool is within the depth of field of the side edge camera lens), the position of the side edge camera structure is limited by the clamp.

The cleaning module is replaced by a machine tool cleaning module, which is used for cleaning the tool with cutting fluid, reduces cutting chips, and ensures the validity of the collected images.

The control module includes an industrial computer, an encoder, a motion control card, and a servo driver; the servo driver is used to drive the push rod motor and the servo motor; the absolute encoder is used to form a feedback system with the servo motor to improve the bottom edge camera Rotation positioning accuracy; motion control card is used to receive generation and storage instructions; industrial computer is used for data calculation. The speed of the servo motor is controlled by the fuzzy PID control method, so that the system can achieve a stable and fast response, and has strong anti-interference ability. And use the S-curve algorithm to plan the trajectory of the input position signal, so that the motor can achieve trapezoidal acceleration and deceleration, and the operation is more stable.

Referring to Fig. 7-Fig. 15, a tool wear detection method based on a tool detection device designed in the present invention is characterized in that it includes two processes of parameter acquisition and wear measurement. The acquisition of parameters takes place before all processing activities start. When the information of a new tool is recorded in the tool magazine, the wear detection device needs to record the information of the tool. This information includes the model of the tool, number of flutes, diameter and image template. Wear measurement means that when the tool is used to process the workpiece, it returns to the tool magazine and before the next processing activity, the tool is taken out by the mechanical arm for wear detection.

The parameter acquisition is characterized in that the process includes the following steps: the following steps are based on milling cutter tool wear detection parameter acquisition steps, and other types of tool initialization steps are taken as an example.

Step 1. Calibration of camera parameters: Calibrate the bottom edge camera 7-3 and the side edge camera 8-3 respectively:

Place the bottom edge camera parameter calibration plate in the depth of field of the bottom edge camera through the tooling, so that the bottom edge camera parameter calibration plate is facing the bottom edge camera 7-3; turn on the bottom edge ring light source 7-1, and move the bottom edge ring light source 7 up and down -1, provide good light for the bottom edge camera 7-3, and judge the rationality of the position of the bottom edge ring light source 7-1 according to the clarity of the imaging inside the camera. Use the bottom edge camera 7-3 to collect 3 photos of the bottom edge camera parameter calibration board, select the clearest photo for camera correction, obtain the internal parameter value of the bottom edge camera 7-3, and eliminate the error caused by lens distortion;

Place the side edge camera parameter calibration plate in the depth of field of the side edge camera through the tooling, make the side edge camera parameter calibration plate face the side edge camera, turn on the side edge ring light source 8-1, and move the side edge ring light source 8-1 up and down, Provide good lighting for the side edge camera 8-3, and judge the rationality of the position of the side edge ring light source 8-1 according to the clarity of the imaging inside the camera. Use the side edge camera 8-3 to collect 3 photos of the side edge camera parameter calibration board, select the clearest photo for camera correction, obtain the internal parameter value of the side edge camera 8-3, and eliminate the error caused by lens distortion;

Step 2. Initialize the position of the image acquisition device: firstly, by adjusting the position of the slider on the guide rail, move the side edge image acquisition device 8 to the farthest position from the center line of the bottom edge camera; Control the operation of the vertical drive electric cylinder 6-2, lower the support platform 6-4 to the lowest position, thereby driving the bottom edge image acquisition device 7 and the side edge image acquisition device 8 to drop to the lowest position;

Simultaneously by controlling the operation of the rotary drive device 6-5 through the control module, the side edge image acquisition device 8 is rotated to the initial position. initial position.

Step 3. Installation of the tool wear detection device: use the tool magazine and the tool changer to transport a standard tool to the tool detection point, and place the tool wear detection device directly below the tool detection point.

Step 4, preliminary adjustment of the image acquisition device: adjust the height of the bottom edge image acquisition device 7 and the side edge image acquisition device 8 by vertically driving the electric cylinder 6-2, so that the bottom edge of the tool is within the depth of field of the bottom edge camera lens. After the position of the bottom edge camera is adjusted in place, turn on the clamp switch in the guide rail pair, and adjust the distance between the side edge camera 8-3 and the tool centerline by moving the slider, so that the side edge of the tool is within the depth of field range of the side edge camera 8-3 , and then close the clamp switch, thereby fixing the position of the side edge camera 8-3.

Step 5. Calibration of pixel equivalent: Obtain the image of the bottom edge of the unworn tool through the bottom edge camera 7-3, and preprocess the image. Here, the image preprocessing process is as follows: the image is grayscaled and filtered successively, and then the contrast of the image is adjusted to complete the image preprocessing.

After the image is preprocessed, the diameter of the tool in the image is extracted by the Hough circle detection method. Assuming that the size of the tool diameter in the image is N pixels, and the actual size of the tool diameter is D, the pixel equivalent is

Step 6. Calculation of the relative position of the bottom edge of the tool and the bottom edge camera: the image of the side edge of the tool is taken by the side edge camera 8-3, the image is preprocessed, and the specific position coordinates of the bottom edge of the tool are obtained through image analysis; The method for calculating the specific position coordinates of the bottom edge of the tool by image analysis is as follows: establish a coordinate system σ ₀ (O ₀ X ₀ Y ₀ Z ₀ ) with the center of the camera lens of the bottom edge, and the coordinate system O ₀ is the center of the camera lens of the bottom edge, The direction of the Z ₀ axis is vertically upward, and the Y ₀ axis is parallel to the extension line of the side edge guide rail; in this coordinate system, the center coordinate of the side edge camera lens is O ₁ (X ₁ Y ₁ Z ₁ ), and the center coordinate of the bottom surface of the tool is O ₂ (X ₂ Y ₂ Z ₂ ). Utilize the side edge camera 8-3 to take the tool image, establish a pixel coordinate system σ ₁ (O _P1 -U ₁ V ₁ ) at the pixel origin O _P1 of the image, take the U ₁ axis as a horizontal line, and the V ₁ axis as a vertical line , so that all pixels of the image that do not coincide with the coordinate axis are in the first quadrant of the coordinate axis. And let the coordinate of the center point of the image in the coordinate system σ ₁ be O ₃ (W ₃ H ₃ ). The image is analyzed by the minimum circumscribing rectangle method, and the circumscribing rectangle of the tool image is obtained. The coordinates of the center point of the four sides of the circumscribing rectangle that are parallel to the bottom surface of the tool and closest to the bottom surface of the tool in the coordinate system are O ₂ (W ₂ H ₂ ). From this, the relative position of the tool bottom edge and the bottom edge camera can be obtained: Z=H ₃ −H ₂ +Z ₁ . The specific position coordinates of the tool are stored, and when the tool is measured next time, the data is automatically called to realize the rapid adjustment of the bottom edge camera position.

Step 7. Positioning of the side edge camera: Obtain the image of the bottom edge of the tool through the bottom edge camera 7-3. After the image is preprocessed, the feature point of the tool closest to the center of the tool circle is obtained through image analysis. Line angle Calculate the minimum angle θ between the line and the vertical line of the image. The method for solving the angle θ is to establish a pixel coordinate system σ ₂ (O _P2 -U ₂ V ₂ ) at the pixel origin _OP2 of the image of the tool bottom edge, the U ₂ axis is a horizontal straight line, and the V ₂ axis is a vertical straight line , so that all pixels of the image that do not coincide with the coordinate axis are in the first quadrant of the coordinate axis. The image is processed by the Hough circle algorithm detection algorithm to obtain the outer contour of the tool in the image, which is a circle; the coordinates of the center of the circle on the image are O ₄ (W ₄ H ₄ ). The tool feature points are detected by the sub-pixel level corner point detection method, and different feature points are selected according to different tools, and the feature point closest to the center of the bottom surface of the tool is selected; the opposite feature points are connected to construct a straight line. By constructing a vertical line through the image center point O ₄ points on the image, find the minimum acute angle γ between the vertical line and the line connecting the relative feature point. The minimum angle from the standard position: θ=γ-α, where α is the characteristic angle of the fixed tool.

Let clockwise be positive in this coordinate system. If the angle θ>0°, the servo motor drives the connecting sleeve 6-9 to rotate θ clockwise; if the angle θ≤0°, then rotate θ counterclockwise;

After the side edge image acquisition device 8 is rotated into place, the bottom edge image is acquired again. Image processing bottom edge image and reading angle θ, if θ is in the range of 0±Δδ, it means that the tool side edge positioning has been completed, where Δδ is determined according to the specific accuracy requirements; if θ does not meet the standard, adjust again, if the number of adjustments More than three times, an alarm is issued to notify the staff that the machine is malfunctioning.

Step 8. Manufacture of the tool bottom edge template: After the camera position is adjusted, the bottom edge camera 7-3 is used to acquire the bottom edge image. Firstly, the image is preprocessed to obtain a clear image. Then the image is processed through the Hough circle algorithm detection algorithm to obtain the outer contour of the tool in the image, which is a circle; the center of the circle is Q ₁ . Take Huo circle center Q ₁ as the center to intercept a square with side length X ₁ , and intercept the internal image of the square; the size of X ₁ is equal to the pixel value corresponding to the diameter of the tool in the image; set the image of the blade as the foreground, and the image except the blade as the background . The foreground of the bottom image of the unworn tool in the intercepted image is extracted by OTUS threshold detection, the pixels in the foreground area are set to 255, and the other pixels are set to 0, and the obtained image is stored as the tool bottom edge template.

通过霍夫直线检测识别出图像中存在的直线，并利用角度信息提取出所需的刀具侧刃边缘直线。设刀刃图像为前景，除刀刃外的图像为后景。将未磨损刀具侧刃图像的前景像素置为255，其他部分像素置为0，所得图片存储为该刀具侧刃模板。Step 9. Manufacture of the tool side edge template: After the camera position is adjusted, the side edge image is acquired through the side edge camera 8-3. Firstly, the image is preprocessed to obtain a clear image. Then process the image through canny edge detection to obtain the profile image of the tool side edge. The minimum circumscribed rectangle of the side edge is identified by the minimum circumscribed rectangle algorithm, and the midpoint of the lower side of the minimum circumscribed rectangle is taken as Q ₂ . Based on Q ₂ , the length and width of the tool side edge image after preprocessing are intercepted as X ₂ and Y _{2 ,} respectively. Rectangular, intercepting the internal image of the rectangle; X ₂ is equal to 1/2 of the corresponding pixel value of the tool diameter in the image, Y ₂ =2X ₂ ; in the intercepted image, establish pixel coordinates at the pixel origin O _P3 of the image System σ ₃ (O _P3 -U ₃ V ₃ ), with the U ₃ axis as a horizontal line and the V ₃ axis as a vertical line, so that all pixels of the image that do not coincide with the coordinate axis are in the first quadrant of the coordinate axis; Q ₂ The coordinates of the point in the coordinate system σ ₃ are

The straight line in the image is identified by Hough line detection, and the required straight line of the side edge of the tool is extracted by using the angle information. Let the blade image be the foreground, and the image other than the blade be the background. Set the foreground pixels of the side edge image of the unworn tool to 255, and set the other pixels to 0, and the resulting image is stored as the tool side edge template.

The wear measurement described above is characterized in that the process includes the following steps: the following steps are milling cutter-based tool wear detection steps, and other types of tool wear measurement steps are listed here.

Step 1. Initialization of the device: same as the initialization step 2 above.

Step 2. Arrival of the tool: Use the tool magazine and the tool changing manipulator to transport a standard tool to the tool inspection point.

Step 3. Position adjustment of the bottom edge camera: the tool magazine obtains the signal of the tool under test, retrieves the stored value Z, and adjusts the height of the bottom edge image acquisition device 7 and the side edge image acquisition device 8 by vertically driving the electric cylinder 6-2; The bottom edge camera 7-3 acquires the image of the bottom edge of the tool, and judges the sharpness of the bottom edge image through the Laplacian gradient method. If the sharpness score is lower than S, it means that the tool magazine has changed the tool incorrectly, and the process is terminated, and the data is fed back; where S Score for clear images based on algorithmic settings.

Step 4. Judgment of the type of tool: After the camera position is adjusted, the image of the bottom edge of the tool is acquired by the bottom edge camera 7-3. Obtain a clear image of the bottom edge of the tool through image preprocessing, and perform template matching with the tool template information in the database to obtain the most matching image, which is the unworn tool template of the tool. If the tool information obtained by image recognition does not match the tool change data in the tool magazine, it means that the tool change in the tool magazine is wrong, the process is terminated, and the data is fed back.

Step 5. Positioning of the side edge camera: the same as the above parameter acquisition step 7, through the obtained θ angle, rotate the side edge image acquisition device 8, so that the side edge camera lens plane and the bottom edge plane of any tool tip and tool center The connection line is vertical; the image of the tool collected by the bottom edge camera is the image of the tool in the standard state.

Step 6. Acquisition of the image of the worn part of the bottom edge: the same as the image of the bottom edge of the tool acquired above. After the image is preprocessed, the Hough circle of the tool is recognized. Take the Hough circle Q ₁ as the center to intercept a square whose side length is X ₁ , and intercept the image inside the square.

Step 7. Image processing of the wear part of the bottom edge: extract the wear area in the inner image of the square through OTUS threshold detection, set the pixels of the wear area to 255, and set the pixels of other parts to 0.

Step 8. Extraction of the wear area of the bottom edge: Mask the worn tool threshold image and the existing unworn tool template, and set the image pixels other than the current knife edge to be extracted to 0;

Step 9. Acquisition of the image of the worn part of the side edge: the same as the above parameter acquisition step 9, a screenshot of the image of the worn part of the side edge of the tool is obtained by image clipping.

Step 10. Image processing of the wear part of the side edge: extract the wear area of the obtained rectangular inner image through OTUS threshold detection, set the pixels of the wear area to 255, and set the pixels of other parts to 0.

Step 11. Extraction of the worn area of the side edge: Mask the worn tool threshold image and the existing unworn tool template, set the image pixels other than the current edge to be extracted to 0; set the worn area as a knife The tip is rotated counterclockwise from the center, and the angle of rotation is the helix angle of the corner cutter, so that the wear area is vertically banded. Repeat several times to extract all flank wear areas.

Step 12, calculating the amount of wear;

For the processed bottom edge and side edge wear images, scan from bottom to top with a horizontal line to calculate the average pixel value a in the horizontal direction of the wear area, and put it into the formula VB=M×a to convert it into the actual average wear The measurement dimension is VB, the bottom edge wear dimension is VB ₁ , and the side edge wear dimension is VB ₂ . Calculates the wear amount of each bottom edge and side edge, and stores the data.

Step 13, wear evaluation; damage characteristic index

If any VB ₁ or VB ₂ of the tool is greater than Lmm, the tool wears excessively, and an alarm is sent to the machining center system to stop using the tool; where L is the bluntness standard set according to the national standard.

Claims (10)

1. A tool detection device that can be integrated on the tool magazine, comprising an image acquisition module and a control module; it is characterized in that, the image acquisition module includes: The bottom edge image acquisition device is used to collect the image of the bottom edge of the tool; The side edge image acquisition device is used to acquire the image of the tool edge measurement; The image acquisition support device can drive the rotation and lifting of the bottom edge image acquisition device and the side edge image acquisition device at the same time; through the lifting, the bottom edge of the tool is located in the depth of field of the lens of the bottom edge image acquisition device, and the side edge image acquisition device obtains the tool The standard position picture of each edge; through rotation, the side edge image acquisition device can obtain the standard position picture of the bottom edge of the tool; The horizontal distance of the side edge image acquisition device relative to the image acquisition support device is adjustable, and is used to adjust the tool side edge to be located within the depth of field of the side edge camera lens; The control module is used to control the rotation and lifting of the image acquisition support device. 2. The tool detection device that can be integrated on the tool magazine according to claim 1, wherein the image acquisition support device includes a fixed frame, a vertical drive electric cylinder, a guide device, a support platform, a rotary drive device, a small Gears, large gears, hollow bushings, connecting bushings, encoders for collecting the position of the connecting bushings and encoders for collecting the position of the vertically driven electric cylinder; The rotary driving device is fixed below the supporting platform, and its driving end is fixedly connected to the pinion; the pinion and the bull gear form a gear transmission, and the bull gear is rotatably connected to the hollow sleeve, and the hollow shaft The sleeve is fixedly connected above the support platform; the connecting shaft sleeve is fixedly connected to the gear wheel; the encoder is fixedly connected to obtain the rotation angle of the connecting shaft sleeve. 3. The tool detection device that can be integrated on the tool magazine according to claim 2, wherein the bottom edge image acquisition device includes a bottom edge ring light source, a bottom edge light source support component, a bottom edge camera, and a bottom edge camera The supporting part; the supporting part of the bottom edge camera is fixedly connected to the lower part of the connecting shaft sleeve, the bottom cutting edge camera is fixedly connected to the supporting part of the bottom cutting edge camera, the supporting part of the bottom cutting edge light source is fixedly connected to the connecting shaft sleeve, and the ring light source of the bottom cutting edge It is fixed on the supporting part of the bottom edge light source to provide light source for the bottom edge camera. 4. The tool detection device that can be integrated on the tool magazine according to claim 2, wherein the side edge image acquisition device includes a side edge ring light source, a side edge light source support, a side edge camera and a side edge camera Support piece; the side edge ring light source is fixedly connected with the side edge camera through the side edge light source supporting part to provide light source for the side edge camera; the side edge camera support is fixedly connected to the connecting sleeve; the side edge camera It is fixed on the side edge camera support, and can slide relative to the side edge camera support, so that the horizontal distance relative to the image acquisition support device can be adjusted. 5. The tool detection device that can be integrated on the tool magazine according to claim 4, wherein the side edge camera support includes a guide rail mounting plate, a guide rail pair containing a clamp and a support rod; the guide rail installation The plate is fixed on the connecting shaft sleeve, the guide rail pair containing the clamp is fixed on the guide rail mounting plate, the support rod is fixed on the slider of the guide rail pair containing the clamp; the side edge camera is fixed on the support rod . 6. The tool wear detection method of the tool detection device that can be integrated on the tool magazine according to any one of claims 1-5, characterized in that it includes two processes of parameter acquisition and wear measurement; The parameter acquisition process is used to enter the information of the new tool before the processing activity starts, including the model of the tool, the number of edges, the diameter and the image template; The wear measurement process is used to return the tool to the tool magazine after processing the workpiece and take out the tool through the mechanical arm for wear detection before the next processing activity. Mask the worn tool threshold image and the unworn image template to perform The extraction of the wear area calculates the average pixel value of the wear area and converts it into the actual average wear size. 7. The tool wear detection method according to claim 6, wherein the parameter process comprises the following steps: Step 1. Calibrate the bottom edge image acquisition device and the side edge image acquisition device respectively; Step 2, initializing the position of the image acquisition device; Step 3. Place the tool wear detection device directly below the tool detection point; Step 4, make the bottom edge of the tool within the depth of field range of the lens of the bottom edge image acquisition device, and make the side edge of the tool within the depth of field range of the side edge image acquisition device; Step 5, acquiring the image of the bottom edge of the unworn tool through the bottom edge image acquisition device; Step 6. Obtain the image of the unworn side edge through the side edge image acquisition device, and obtain the position Z of the bottom edge of the tool relative to the bottom edge image acquisition device through image analysis; Step 7. Positioning of the side edge camera: through image analysis, the feature point of the tool closest to the center of the tool circle is obtained, and the minimum angle θ between the line and the vertical line of the image is obtained according to the angle of the line formed by the line connecting the feature points; when the side edge image After the acquisition device is rotated in place, the image of the bottom edge is acquired again, and the angle θ is read. If θ is within the set range, it means that the positioning of the side edge of the tool has been completed; Step 8. Manufacture of tool template: Obtain the bottom edge image through the bottom edge image acquisition device, process the image through the Hough circle algorithm detection algorithm to obtain the outer contour of the tool in the image, take the center of the Huo circle Q ₁ as the center to intercept a square with a side length of X ₁ , and intercept the internal image of the square; X ₁ The size is equal to the pixel value corresponding to the diameter of the tool in the image; the image of the blade is set as the foreground, and the image other than the blade is the background; the foreground of the bottom image of the unworn tool in the intercepted image is extracted through OTUS threshold detection, and the pixels in the foreground area are Set it to 255, set other pixels to 0, and store the resulting image as the tool bottom edge template; The side edge image is acquired by the side edge image acquisition device, the tool side edge contour image is obtained by canny edge detection and image processing, and the minimum circumscribed rectangle of the side edge is identified by the minimum circumscribed rectangle algorithm. The midpoint of the lower side of the minimum circumscribed rectangle is taken as Q ₂ , and Q ₂ is based on the preprocessed tool side edge image to intercept a rectangle whose length and width are X ₂ and Y ₂ respectively, and intercept the internal image of the rectangle; the size of X ₂ is equal to 1/2 of the pixel value corresponding to the tool diameter in the image , Y ₂ =2X ₂ ; in the intercepted image, establish a pixel coordinate system σ ₃ (OP ₃ -U ₃ V ₃ ) at the pixel origin OP ₃ of the image, with the U ₃ axis as the horizontal line and the V ₃ axis as the vertical A straight line, so that all the pixels of the image that do not coincide with the coordinate axis are in the first quadrant of the coordinate axis; the coordinates of point Q ₂ in the coordinate system σ ₃ are

The straight line in the image is identified by Hough line detection, and the required edge line of the tool side edge is extracted by using the angle information; the knife edge image is set as the foreground, and the image except the knife edge is the background; the side edge image of the unworn tool Set the foreground pixels to 255, and set the other pixels to 0, and store the resulting picture as the tool side edge template. 8. The tool wear detection method according to claim 7, wherein the wear measurement comprises the following steps: Step 9, position initialization of the image acquisition device; Step 10, using the tool magazine and the tool changing manipulator to transport a standard tool to the tool inspection point; Step 11. Position adjustment of the bottom edge camera: The tool magazine obtains the signal of the tool under test, retrieves the position Z value, adjusts the bottom edge image acquisition device and the side edge image acquisition device, acquires the image of the bottom edge of the tool, and uses the Laplacian gradient method Judging the sharpness of the bottom edge image, if the sharpness score is lower than the set value, it means that the tool changer is wrong, and the process is terminated; Step 12. Judgment of the type of tool: Obtain the image of the bottom edge of the tool, perform template matching with the tool template information in the database, and obtain the most matching image, which is the unworn tool template of the tool; if the image recognition If the obtained tool information does not match the tool change data in the tool magazine, it means that the tool change in the tool magazine is wrong, and the process is terminated; Step 13. Positioning of the side edge camera: the same as step 7 of the parameter acquisition process, so that the plane of the side edge camera lens is perpendicular to the line connecting the tool tip and the center of the tool on the plane of the bottom edge of any tool; Step 14. Acquisition of the image of the worn part of the bottom edge: After the image is preprocessed, the Hough circle of the tool is recognized, and a square with a side length of X ₁ is intercepted with the Hough circle Q ₁ as the center, and the internal image of the square is intercepted; Step 15. Image processing of the worn part of the bottom edge: extract the worn area in the inner image of the square through OTUS threshold detection, set the pixels of the worn area to 255, and set the pixels of other parts to 0; Step 16. Extraction of the worn area of the bottom edge: Mask the worn tool threshold image and the existing unworn tool template, and set the image pixels other than the current knife edge to be extracted to 0; Step 17. Acquisition of the image of the wear part of the side edge: obtain a screenshot of the image of the wear part of the side edge of the tool through image clipping; Step 18. Image processing of the worn part of the side edge: extract the worn area of the obtained rectangular inner image through OTUS threshold detection, set the pixels of the worn area to 255, and set the pixels of other parts to 0; Step 19. Extraction of the worn area of the side edge: Mask the worn tool threshold image and the existing unworn tool template, set the image pixels other than the current edge to be extracted to 0; set the worn area as a knife Rotate counterclockwise with the tip as the center, and the angle of rotation is the helix angle of the corner cutter, so that the wear area is vertically banded; repeat several times to extract all side edge wear areas; Step 12. Calculate the wear amount; for the processed bottom edge and side edge wear images, scan from bottom to top through a horizontal straight line, calculate the average pixel value in the horizontal direction of the wear area, and convert it into the actual average wear size . 9. The tool wear detection method according to claim 7, characterized in that, the calculation process of the position of the tool bottom edge relative to the bottom edge image acquisition device is: establish a coordinate system with the center of the lens of the bottom edge image acquisition device σ ₀ (O ₀ -X ₀ Y ₀ Z ₀ ), the origin O ₀ is the center of the bottom edge camera lens, the Z ₀ axis is vertically upward, and the Y ₀ axis is parallel to the extension line of the side edge guide rail; in this coordinate system, set The center coordinate of the lens of the side edge image acquisition device is O ₁ (X ₁ Y ₁ Z ₁ ), and the center coordinate of the bottom surface of the tool is O ₂ (X ₂ Y ₂ Z ₂ ); The pixel origin O _p1 establishes a pixel coordinate system σ ₁ (O _P1 -U ₁ V ₁ ), with the U ₁ axis as a horizontal line and the V ₁ axis as a vertical line, so that all pixel points in the image that do not coincide with the coordinate axis are at this coordinate The first quadrant of the axis; and the coordinates of the center point of the image in the coordinate system σ ₁ are O ₃ (W ₃ H ₃ ); the image is analyzed by the minimum circumscribed rectangle method, and the circumscribed rectangle of the tool image is obtained, and the four sides of the circumscribed rectangle are set The coordinates of the center point of the side parallel to the bottom surface of the tool and closest to the bottom surface of the tool in this coordinate system are O ₂ (W ₂ H ₂ ), then the relative position of the bottom edge of the tool and the camera: Z＝H ₃ -H ₂ Z ₁ . 10. The tool wear detection method according to claim 7, wherein the method for solving the angle θ is: establishing a pixel coordinate system σ ₂ (O _P2 −U ₂ V ₂ at the pixel origin O _P2 of the image of the bottom edge of the tool ), the U ₂ axis is a horizontal straight line, and the V ₂ axis is a vertical straight line, so that all pixels of the image that do not coincide with the coordinate axis are in the first quadrant of the coordinate axis; the image is processed by the Hough circle algorithm detection algorithm to obtain the outside of the tool in the image Outline, the outline is a circle; the coordinates of the center of the circle on the image are O ₄ (W ₄ H ₄ ); the tool feature points are detected by the sub-pixel level corner point detection method, and different feature points are selected according to different tools. Select the feature point closest to the center of the bottom surface of the tool; connect the relative feature points to construct a straight line; construct a vertical line through the image center point O ₄ points on the image, and find the minimum acute angle between the vertical line and the line connecting the relative feature points γ; the minimum angle from the standard position: θ=γ-α, where α is the characteristic angle of the fixed tool.

Images