A grinding wheel dressing profile monitoring method based on monocular vision sensor

Abstract

The application provides a grinding wheel dressing profile monitoring method based on a monocular vision sensor, comprising the following steps: introducing a standard profile image of a grinding wheel; collecting a profile image of a grinding wheel to be dressed; performing image segmentation processing, comparing with the standard profile image, and calculating position coordinates and angles of profile out-of-tolerance; a grinding wheel dressing controller performs feed compensation; an actuator adjusts the feed of the dressing grinding wheel, and after adjustment, the dressing is completed; steps two to five are repeated until the deviation between the profile of the grinding wheel to be dressed and the standard profile meets the profile tolerance threshold range. Through the combination of monocular vision and multi-level image processing algorithm, the detection accuracy is improved, and the error is reduced. Through the combination control of the profile recognition calculator and the grinding wheel dressing controller, the process accuracy such as the size accuracy and the profile accuracy of the grinding wheel dressing is ensured.

Description

Technical Field

[0001] This invention relates to the field of grinding wheel dressing technology, and in particular to a method for monitoring the dressing profile of grinding wheels based on a monocular vision sensor. Background Technology

[0002] As a core auxiliary equipment in the grinding of solid carbide tools, the grinding wheel dressing machine needs to repair micro-defects in the grinding wheel during dressing to ensure the dimensional and contour accuracy of the grinding wheel. This directly affects the processing quality, efficiency, cost, and process stability of the grinding. The verification of the grinding wheel dressing machine and the grinding wheel dressing inspection system directly determines the grinding processing quality and production efficiency.

[0003] Currently, wheel dressing and wheel inspection in the overall cemented carbide tool machining industry are still mostly separate configurations with limited in-machine inspection functions, resulting in significant limitations. The process accuracy and measurement benchmarks are reduced due to issues such as wheel reloading and fixture wear. Frequent disassembly and assembly lead to a decrease in the effective working time, increasing equipment, labor, and time costs, and ultimately reducing production efficiency.

[0004] In the prior art, publication number CN106649930A, entitled "An Online Measurement Method for the Curved Profile of Grinding Wheels," discloses an online detection method using acoustic emission sensors to monitor the grinding wheel dressing process. By using acoustic emission sensors and related acquisition equipment placed near the dresser, acoustic emission signals during the dressing process are acquired in real time. These signals are then processed to obtain the actual dressing depth. This method can estimate the grinding wheel dressing profile online, thus achieving online measurement of the curved profile. However, this method has disadvantages such as poor noise immunity, low detection accuracy, and the need for disassembly after dressing, which can introduce repeatability errors during processing. Summary of the Invention

[0005] This invention addresses the aforementioned problems and aims to provide a method for monitoring the contour of grinding wheel dressing based on a monocular vision sensor. By combining monocular vision with a multi-level image processing algorithm, the detection accuracy is improved and errors are reduced. Through the combined control of a contour recognition calculator and a grinding wheel dressing controller, the dimensional accuracy, contour accuracy, and other process accuracy of grinding wheel dressing are ensured, significantly improving grinding precision and further achieving cost reduction and efficiency improvement in grinding processes.

[0006] Specifically, the first aspect of the present invention provides a method for monitoring the contour of grinding wheel dressing based on a monocular vision sensor, comprising the following steps: Step 1: Import the standard profile image of the grinding wheel; Step 2: Acquire the outline image of the grinding wheel to be dressed using a monocular vision sensor; Step 3: The contour recognition calculator performs image segmentation processing on the grinding wheel contour image acquired by the monocular vision sensor, compares it with the standard contour image, and calculates the position coordinates and angles of the contour deviation. Step 4: The grinding wheel dressing controller compensates for the feed by calculating the out-of-tolerance position angle of the profile using the profile recognition calculator; Step 5: The actuator adjusts the feed of the dressing wheel through the signal output by the dressing controller, and the dressing is completed after adjustment; Step 6: Repeat steps 2 to 5 until the deviation between the profile of the grinding wheel to be dressed and the standard profile meets the profile tolerance threshold range.

[0007] Import the standard profile image of the grinding wheel, specifically import the data image of a pre-saved CAD standard grinding wheel profile model.

[0008] Furthermore, the monocular vision sensor employs a multi-layer lens structure and is equipped with a parallel light source.

[0009] Furthermore, the contour recognition calculator includes an image processing module and a contour deviation calculation module.

[0010] Furthermore, the image processing module performs image segmentation processing and extracts the actual grinding wheel contour based on binarization and edge detection techniques.

[0011] The image processing module first preprocesses the grinding wheel contour image acquired by the monocular vision sensor, including image grayscale processing by weighted averaging, then binarizing the grayscale image using adaptive thresholding to enhance contrast, and then performing multi-level noise reduction, including median filtering and opening operation noise reduction. Finally, it detects the grinding wheel edge based on the Canny edge detection operator and performs image segmentation processing to extract the actual grinding wheel contour.

[0012] Furthermore, the contour deviation calculation module compares the standard contour image with the actual grinding wheel contour extracted by the image processing module to obtain the position coordinates and angles of the features to be repaired in the actual contour.

[0013] Furthermore, the process of the profile deviation calculation module performing deviation comparison is as follows: Step 1.1: Spatial alignment and coordinate system matching between the standard profile image and the actual grinding wheel profile; The geometric centroid of the grinding wheel is selected as the reference point for spatial alignment of the standard profile and the actual grinding wheel profile. The actual profile is aligned with the standard profile through rigid body transformation (translation + rotation). Translation means translating the reference point of the actual profile to the reference point position of the standard profile. Rotation means calculating the angle between the principal directions of the actual profile and the standard profile (such as through principal axis fitting or least squares matching). The actual profile is rotated to make the principal directions coincide. After completion, the standard profile and the actual grinding wheel profile are in the same coordinate system.

[0014] Step 1.2: Based on the shortest Euclidean distance from each point in the actual grinding wheel profile point set to the standard profile point set, generate the distance deviation sequence of the actual profile and set the profile tolerance threshold range; Calculate the shortest Euclidean distance from each point in the actual grinding wheel profile point set to the standard profile point set. If a point in the actual grinding wheel profile is located outside the standard profile, the shortest Euclidean distance is positive, indicating a potential high point; conversely, if a point in the actual grinding wheel profile is located inside the standard profile, the shortest Euclidean distance is negative, indicating a potential collapse. After the calculation is completed, a distance deviation sequence of the actual profile is generated.

[0015] The profile tolerance threshold range setting includes positive and negative thresholds. Deviations within the profile tolerance threshold range are considered normal deviations; otherwise, they are potential adjustment points, i.e. out-of-tolerance points. The profile tolerance threshold range setting is determined according to actual accuracy requirements.

[0016] Step 1.3: Calculate the centroid coordinates for each set of consecutive out-of-tolerance points. With the center of the grinding wheel as the origin, calculate the polar angle of the centroid coordinates and record the central angle, angle range, centroid coordinates and deviation type.

[0017] A set of consecutive out-of-tolerance points must satisfy the condition that the number of consecutive out-of-tolerance points is greater than a minimum threshold (5 ≤ minimum threshold ≤ 10). The region formed by the set of consecutive out-of-tolerance points is a region to be repaired. Calculate the centroid coordinates of the region to be repaired, taking the center of the grinding wheel as the origin, and calculate the polar angle of the centroid coordinates (i.e., the region to be repaired). Record the central angle, angle range, centroid coordinates, and deviation type. The specific characteristics of the region to be repaired are determined as follows: The features to be dressed include wheel collapse, high points, and distortion. The deviation type in step 1.3 corresponds to the type of feature to be dressed. Collapse is generally an edge collapse, that is, a sharp contour (such as the edge of a forming grinding wheel) becomes rounded. When the distance deviation sequence of a continuous set of out-of-tolerance points is all negative, it is judged as a collapse. High points are generally local bulges on the surface of the grinding wheel (such as abrasive grain adhesion blocks, accumulation of unshed grinding debris, or local material expansion). When the distance deviation sequence of a continuous set of out-of-tolerance points is positive, it is judged as a high point. Distortion is caused by long-term grinding and wear, which may result in uneven wear on the surface of the grinding wheel (such as eccentricity, passivation, and blockage). When the distance deviation sequence of the continuous out-of-range point set changes from positive to negative, it is judged as distortion.

[0018] Furthermore, the grinding wheel dressing controller includes a dressing wheel, an actuator, and absolute encoders respectively mounted on the feed Z-axis and the rotation C-axis.

[0019] Furthermore, the absolute encoder is used to provide feedback on the absolute position of the dressing wheel on the feed Z-axis and rotation C-axis.

[0020] The absolute encoder of the grinding wheel dressing controller is arranged on the Z and C axes. Based on the position coordinates and angles of features such as collapse, high points, and undressed areas obtained from the contour recognition calculator, the absolute positions of the Z and C axes are calculated.

[0021] Furthermore, the actuator is used to control and adjust the position of the dressing wheel on the feed Z-axis and rotation C-axis.

[0022] Furthermore, the movement of the dressing wheel includes: reciprocating linear motion in the direction of the S-axis of the dressing wheel spindle, linear motion in the direction of the feed Z-axis, and rotational motion with the rotation C-axis as the center.

[0023] The polar angle, central angle, angle range, and centroid coordinates of the area to be corrected are used to determine the size of the area to be corrected. The specific correction actions corresponding to different deviation types are as follows: When the deviation type is collapse, the repair goal is to restore sharpness. The collapse depth is the magnitude of the negative value of the continuous out-of-tolerance points. The standard profile needs to be shifted downwards. The shift value is the maximum value of the collapse depth. The profile out-of-tolerance is recalculated, converted to the deviation type of high point, and the protruding part on the actual profile is removed to obtain a consistent grinding wheel profile. When the deviation type is high point, the repair target is to remove the protrusion. The Z-axis is directly cut in (depth = protrusion height), and the C-axis is cut at high speed (to avoid adhesion and diffusion). The protrusion height is the magnitude of the positive value of the continuous out-of-tolerance points. When the deviation type is distortion, the correction target is to reshape a uniform contour, with Z-axis sinusoidal feed (matching the ripple depth), C-axis uniform rotation + S-axis synchronous reciprocating. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of this drawing or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this drawing. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0025] Figure 1 This is a flowchart of the steps of the present invention; Figure 2 This is a schematic diagram illustrating the collapse repair process according to the present invention; Figure 3 This is an enlarged schematic diagram of area A showing the collapse repair situation according to the present invention; Figure 4 This is a schematic diagram of the actual repaired portion of the collapse repair process according to the present invention.

[0026] The purpose, features, and advantages of this accompanying drawing will be further explained in conjunction with the embodiments and with reference to the accompanying drawing. Detailed Implementation

[0027] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be described and illustrated below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. All other embodiments obtained by those skilled in the art based on the embodiments provided by this invention without inventive effort are within the scope of protection of this invention.

[0028] Obviously, the accompanying drawings described below are merely some examples or embodiments of the present invention. Those skilled in the art can apply the present invention to other similar scenarios based on these drawings without any inventive effort. Furthermore, it is understood that although the efforts made in this development process may be complex and lengthy, for those skilled in the art related to the content disclosed in this invention, modifications to design, manufacturing, or production based on the technical content disclosed in this invention are merely conventional technical means and should not be construed as insufficient disclosure of the present invention.

[0029] Unless otherwise specified, all embodiments and optional embodiments of the present invention can be combined with each other to form new technical solutions.

[0030] Unless otherwise specified, all technical features and optional technical features of this invention can be combined to form new technical solutions.

[0031] Unless otherwise specified, all steps of the present invention may be performed sequentially or randomly, preferably sequentially. For example, the method includes steps (a) and (b), indicating that the method may include steps (a) and (b) performed sequentially, or it may include steps (b) and (a) performed sequentially. For example, the mention that the method may also include step (c) indicates that step (c) may be added to the method in any order. For example, the method may include steps (a), (b), and (c), or it may include steps (a), (c), and (b), or it may include steps (c), (a), and (b), etc.

[0032] Unless otherwise specified, the terms "comprising" and "including" as used in this invention can be open-ended or closed-ended. For example, "comprising" and "including" can mean that other components not listed may also be included, or that only the listed components may be included.

[0033] Unless otherwise specified, the term "or" is inclusive in this invention. For example, the phrase "A or B" means "A, B, or both A and B". More specifically, the condition "A or B" is satisfied by any of the following conditions: A is true (or exists) and B is false (or does not exist); A is false (or does not exist) and B is true (or exists); or both A and B are true (or exist).

[0034] To better understand the solutions of the embodiments of the present invention, some related terms and concepts that may be involved in the embodiments of the present invention will be introduced below.

[0035] (1) A monocular vision sensor is a system that uses only one camera for image capture and processing. It is widely used in smart devices and monitoring systems, but its main challenge lies in the inability to directly acquire depth information, thus requiring algorithms to infer the 3D structure of the scene. Compared with binocular vision, monocular vision has certain limitations in depth estimation, but it still plays an important role in fields such as autonomous driving and object detection. Geometric constraints and deep neural networks can be used to assist monocular vision systems in 3D perception.

[0036] (2) Adaptive Thresholding is an adaptive local thresholding segmentation algorithm, which is a binarization method that dynamically calculates the threshold based on the brightness distribution of local regions of the image. For each pixel, the threshold is calculated separately based on the statistical characteristics (mean, Gaussian weighted sum or median) of its neighborhood (such as a window with radius r).

[0037] (3) Canny Edge Detector is a multi-stage optimal edge detection algorithm that includes four core steps of gradient edge detection, namely Gaussian filtering (smoothing and denoising the image), gradient calculation (using the Sobel operator to calculate the gradient magnitude), non-maximum suppression (preserving local maximum points in the gradient direction), and double threshold hysteresis (high threshold determines strong edges, and low threshold connects weak edges).

[0038] In this embodiment, as Figure 1 As shown, a method for monitoring the contour of grinding wheel dressing based on a monocular vision sensor includes the following steps: Step 1: Import the standard profile image of the grinding wheel; Step 2: Acquire the outline image of the grinding wheel to be dressed using a monocular vision sensor; Step 3: The contour recognition calculator performs image segmentation processing on the grinding wheel contour image acquired by the monocular vision sensor, compares it with the standard contour image, and calculates the position coordinates and angles of the contour deviation. Step 4: The grinding wheel dressing controller compensates for the feed by calculating the out-of-tolerance position angle of the profile using the profile recognition calculator; Step 5: The actuator adjusts the feed of the dressing wheel through the signal output by the dressing controller, and the dressing is completed after adjustment; Step 6: Repeat steps 2 to 5 until the deviation between the profile of the grinding wheel to be dressed and the standard profile meets the profile tolerance threshold range.

[0039] Import the standard profile image of the grinding wheel, specifically import the data image of a pre-saved CAD standard grinding wheel profile model.

[0040] Furthermore, the monocular vision sensor employs a multi-layer lens structure and is equipped with a parallel light source.

[0041] Furthermore, the contour recognition calculator includes an image processing module and a contour deviation calculation module.

[0042] Furthermore, the image processing module performs image segmentation processing and extracts the actual grinding wheel contour based on binarization and edge detection techniques.

[0043] The image processing module first preprocesses the grinding wheel contour image acquired by the monocular vision sensor, including image grayscale processing by weighted averaging, then binarizing the grayscale image using adaptive thresholding to enhance contrast, and then performing multi-level noise reduction, including median filtering and opening operation noise reduction. Finally, it detects the grinding wheel edge based on the Canny edge detection operator and performs image segmentation processing to extract the actual grinding wheel contour.

[0044] Furthermore, the contour deviation calculation module compares the standard contour image with the actual grinding wheel contour extracted by the image processing module to obtain the position coordinates and angles of the features to be repaired in the actual contour.

[0045] Furthermore, the process of the profile deviation calculation module performing deviation comparison is as follows: Step 1.1: Spatial alignment and coordinate system matching between the standard profile image and the actual grinding wheel profile; The geometric centroid of the grinding wheel is selected as the reference point for spatial alignment of the standard profile and the actual grinding wheel profile. The actual profile is aligned with the standard profile through rigid body transformation (translation + rotation). Translation means translating the reference point of the actual profile to the reference point position of the standard profile. Rotation means calculating the angle between the principal directions of the actual profile and the standard profile (such as through principal axis fitting or least squares matching). The actual profile is rotated to make the principal directions coincide. After completion, the standard profile and the actual grinding wheel profile are in the same coordinate system.

[0046] Step 1.2: Based on the shortest Euclidean distance from each point in the actual grinding wheel profile point set to the standard profile point set, generate the distance deviation sequence of the actual profile and set the profile tolerance threshold range; Calculate the shortest Euclidean distance from each point in the actual grinding wheel profile point set to the standard profile point set. If a point in the actual grinding wheel profile is located outside the standard profile, the shortest Euclidean distance is positive, indicating a potential high point; conversely, if a point in the actual grinding wheel profile is located inside the standard profile, the shortest Euclidean distance is negative, indicating a potential collapse. After the calculation is completed, a distance deviation sequence of the actual profile is generated.

[0047] The profile tolerance threshold range setting includes positive and negative thresholds. Deviations within the profile tolerance threshold range are considered normal deviations; otherwise, they are potential adjustment points, i.e. out-of-tolerance points. The profile tolerance threshold range setting is determined according to actual accuracy requirements.

[0048] Step 1.3: Calculate the centroid coordinates for each set of consecutive out-of-tolerance points. With the center of the grinding wheel as the origin, calculate the polar angle of the centroid coordinates and record the central angle, angle range, centroid coordinates and deviation type.

[0049] A set of consecutive out-of-tolerance points must satisfy the condition that the number of consecutive out-of-tolerance points is greater than a minimum threshold (5 ≤ minimum threshold ≤ 10). The region formed by the set of consecutive out-of-tolerance points is a region to be repaired. Calculate the centroid coordinates of the region to be repaired, taking the center of the grinding wheel as the origin, and calculate the polar angle of the centroid coordinates (i.e., the region to be repaired). Record the central angle, angle range, centroid coordinates, and deviation type. The specific characteristics of the region to be repaired are determined as follows: The features to be dressed include wheel collapse, high points, and distortion. The deviation type in step 1.3 corresponds to the type of feature to be dressed. like Figure 2 As shown in the figure, the black area represents the standard contour, the yellow area represents the actual contour, and the green area represents the area to be repaired. Collapse generally refers to edge collapse, that is, sharp contours (such as the edges of a forming grinding wheel) become rounded. Collapse is judged when the distance deviation sequence of a continuous set of out-of-tolerance points is all negative. Figure 2 An enlarged schematic diagram of part A is as follows Figure 3 As shown.

[0050] High points are generally local bulges on the surface of the grinding wheel (such as abrasive grain adhesion blocks, accumulation of unshed grinding debris, or local material expansion). When the distance deviation sequence of a continuous set of out-of-tolerance points is positive, it is judged as a high point. Distortion is caused by long-term grinding and wear, which may result in uneven wear on the surface of the grinding wheel (such as eccentricity, passivation, and blockage). When the distance deviation sequence of the continuous out-of-range point set changes from positive to negative, it is judged as distortion.

[0051] Furthermore, the grinding wheel dressing controller includes a dressing wheel, an actuator, and absolute encoders respectively mounted on the feed Z-axis and the rotation C-axis.

[0052] Furthermore, the absolute encoder is used to provide feedback on the absolute position of the dressing wheel on the feed Z-axis and rotation C-axis.

[0053] The absolute encoder of the grinding wheel dressing controller is arranged on the Z and C axes. Based on the position coordinates and angles of features such as collapse, high points, and undressed areas obtained from the contour recognition calculator, the absolute positions of the Z and C axes are calculated.

[0054] Furthermore, the actuator is used to control and adjust the position of the dressing wheel on the feed Z-axis and rotation C-axis.

[0055] Furthermore, the movement of the dressing wheel includes: reciprocating linear motion in the direction of the S-axis of the dressing wheel spindle, linear motion in the direction of the feed Z-axis, and rotational motion with the rotation C-axis as the center.

[0056] The polar angle, central angle, angle range, and centroid coordinates of the area to be corrected are used to determine the size of the area to be corrected. The specific correction actions corresponding to different deviation types are as follows: When the deviation type is collapse, the repair goal is to restore sharpness. The collapse depth is the magnitude of the negative value of the continuous out-of-tolerance points. The standard profile needs to be shifted downwards. The shift value is the maximum value of the collapse depth. The profile out-of-tolerance is recalculated, converted to the deviation type of high point, and the protruding part on the actual profile is removed to obtain a consistent grinding wheel profile. right Figure 2 When repairing the collapse defect, the comparison image after translation is as follows: Figure 4 As shown in the figure, the black part is the standard outline, the yellow part is the actual outline, and the blue-green grid part is the area to be trimmed and needs to be cut off.

[0057] When the deviation type is high point, the repair target is to remove the protrusion. The Z-axis is directly cut in (depth = protrusion height), and the C-axis is cut at high speed (to avoid adhesion and diffusion). The protrusion height is the magnitude of the positive value of the continuous out-of-tolerance points. When the deviation type is distortion, the correction target is to reshape a uniform contour, with Z-axis sinusoidal feed (matching the ripple depth), C-axis uniform rotation + S-axis synchronous reciprocating.

[0058] It should be noted that the present invention is not limited to the above-described embodiments. The above embodiments are merely examples, and any embodiments that have the same structure and perform the same effects as the technical concept within the scope of the present invention are included within the scope of the present invention. Furthermore, various modifications that can be conceived by those skilled in the art to the embodiments, and other ways of constructing by combining some of the constituent elements of the embodiments, without departing from the spirit of the present invention, are also included within the scope of the present invention.

Claims

1. A method for monitoring the contour of grinding wheel dressing based on a monocular vision sensor, characterized in that, Includes the following steps: Step 1: Import the standard profile image of the grinding wheel; Step 2: Acquire the outline image of the grinding wheel to be dressed using a monocular vision sensor; Step 3: The contour recognition calculator performs image segmentation processing on the grinding wheel contour image acquired by the monocular vision sensor, compares it with the standard contour image, and calculates the position coordinates and angles of the contour deviation. Step 4: The grinding wheel dressing controller compensates for the feed by calculating the out-of-tolerance position angle of the profile using the profile recognition calculator; Step 5: The actuator adjusts the feed of the dressing wheel through the signal output by the dressing controller, and the dressing is completed after adjustment; Step 6: Repeat steps 2 to 5 until the deviation between the profile of the grinding wheel to be dressed and the standard profile meets the profile tolerance threshold range; The contour recognition calculator includes an image processing module and a contour deviation calculation module. The image processing module is used to perform image segmentation processing and extract the actual grinding wheel contour based on binarization and edge detection technology; The contour deviation calculation module is used to obtain the position coordinates and angles of the features to be repaired in the actual contour by comparing the standard contour image with the actual grinding wheel contour extracted by the image processing module.

2. The method for monitoring the contour of grinding wheel dressing based on a monocular vision sensor according to claim 1, characterized in that, The monocular vision sensor employs a multi-layer lens structure and is equipped with a parallel light source.

3. The method for monitoring the contour of grinding wheel dressing based on a monocular vision sensor according to claim 1, characterized in that, The process of the profile deviation calculation module performing deviation comparison is as follows: Step 1.1: Spatial alignment and coordinate system matching between the standard profile image and the actual grinding wheel profile; Step 1.2: Based on the shortest Euclidean distance from each point in the actual grinding wheel profile point set to the standard profile point set, generate the distance deviation sequence of the actual profile and set the profile tolerance threshold range; Step 1.3: Calculate the centroid coordinates for each set of consecutive out-of-tolerance points. With the center of the grinding wheel as the origin, calculate the polar angle of the centroid coordinates and record the central angle, angle range, centroid coordinates and deviation type.

4. The method for monitoring the contour of grinding wheel dressing based on a monocular vision sensor according to claim 1, characterized in that, The grinding wheel dressing controller includes a dressing wheel, an actuator, and absolute encoders respectively mounted on the feed Z-axis and the rotation C-axis.

5. The method for monitoring the contour of grinding wheel dressing based on a monocular vision sensor according to claim 4, characterized in that, The absolute encoder is used to provide feedback on the absolute position of the dressing wheel on the feed Z-axis and rotation C-axis.

6. The method for monitoring the contour of grinding wheel dressing based on a monocular vision sensor according to claim 4, characterized in that, The actuator is used to control and adjust the position of the dressing wheel on the feed Z-axis and rotation C-axis.

7. The method for monitoring the contour of grinding wheel dressing based on a monocular vision sensor according to claim 4, characterized in that, The movement modes of the dressing wheel include: reciprocating linear motion along the S-axis of the dressing wheel spindle, linear motion along the Z-axis of the feed direction, and rotational motion with the C-axis as the center.

Images