Ceramic chopper size grinding method and device based on image analysis and storage medium
Through image analysis technology, the outer contour and grinding wheel corner information are obtained, and the grinding path is generated, which solves the problem of low grinding accuracy and efficiency of traditional ceramic chopping knives, and achieves high-precision and efficient ceramic chopping knives.
Patent Information
- Application Number
- CN202411073580.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-06
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2044-08-06
AI Technical Summary
The traditional ceramic chopping knife grinding method has problems of low accuracy and efficiency, which relies on the experience and feeling of workers, resulting in uneven polishing and inaccurate size.
Using an image analysis method, by collecting the surface image of the ceramic splitter knife, using the image processing model to obtain the outer contour information and the grinding wheel corner information, generate grinding path information, including grinding direction, force and angle, and control the grinding wheel to perform the ceramic splitter knife. Precise polishing.
The grinding accuracy and efficiency of the ceramic chopping knife are improved, ensuring that the shape and size of the ceramic chopping knife meet the expected standards, and reducing artificial errors.
Smart Images

Figure CN118905741A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of mechanical processing technology, and in particular to a ceramic splitting cutter size grinding method, equipment and storage medium based on image analysis. Background Art
[0002] During the use of ceramic splitters, due to factors such as wear, their size and shape may change, and they need to be regularly polished to restore their accuracy. Traditional polishing methods often rely on the experience and feeling of workers, and there are problems such as uneven polishing and inaccurate size.
[0003] Therefore, how to improve the accuracy and efficiency of grinding ceramic splitters is an urgent problem to be solved. Summary of the invention
[0004] The present application provides a ceramic splitting knife size grinding method, equipment and storage medium based on image analysis, which are used to solve the problem of low precision and efficiency in grinding ceramic splitting knives.
[0005] In a first aspect, the present application provides a ceramic splitting tool size grinding method based on image analysis, comprising: Acquiring a surface image of a ceramic splitting knife to be sharpened; Analyzing the surface image through an image processing model to obtain outer contour information of the ceramic splitting knife, wherein the outer contour information includes a corner point of the ceramic splitting knife, a head length of the ceramic splitting knife, a side length of the ceramic splitting knife, and a side angle of the ceramic splitting knife; Acquire the grinding wheel corner point information of the grinding wheel tool for grinding the ceramic splitter; Generate grinding path information for grinding the ceramic splitting knife according to the outer contour information and the grinding wheel corner point information, wherein the grinding path information includes a grinding direction, a grinding force, and a grinding angle; The grinding wheel tool is controlled to grind the ceramic splitting knife according to the grinding path information.
[0006] Optionally, analyzing the surface image by an image processing model to obtain the outer contour information of the ceramic splitting knife includes: Obtaining color parameters and preset color weights of the surface image; converting the surface image into a grayscale image of the surface image according to the color parameter and the preset color weight; Performing normalization and denoising processing on the grayscale image to generate a processed first image; Inputting the first image into the pre-trained image processing model, and extracting the shape features of the ceramic splitting knife from the first image through the image processing model; the image processing model includes an input layer, a convolution layer, a pooling layer, a fully connected layer, and an output layer, the convolution layer is used to extract local features in the first image, the local features include edge features and corner features, and the pooling layer is used to reduce the dimension of the local features by downsampling; According to the extracted shape features of the ceramic splitting knife, the head length, side length and side angle of the ceramic splitting knife are calculated and generated; Generate the outer contour information according to the head length, side length and side angle of the ceramic splitting knife; The step of calculating and generating the head length, side length and side angle of the ceramic splitting knife according to the extracted shape features of the ceramic splitting knife comprises: Extracting a contour point set of the ceramic splitting knife according to the shape feature of the ceramic splitting knife, wherein the contour point set represents the outer boundary of the ceramic splitting knife; Matching the contour point set and the ceramic splitting knife feature region template by a shape matching algorithm to obtain the similarity between the contour point set and the ceramic splitting knife feature region template, wherein the ceramic splitting knife feature region template is a template pre-drawn according to the type and model of the ceramic splitting knife, and the ceramic splitting knife feature region template includes a head region and a side region; According to the similarity, determining a first subset of contour points corresponding to the head region of the ceramic splitting knife and a second subset of contour points corresponding to the side region of the ceramic splitting knife; Calculating the length of the contour line segment in the head region by using the first contour point subset to obtain the head length of the ceramic splitting knife; Calculating the length of the contour line segment in the side area by using the second contour point subset to obtain the side length of the ceramic splitting knife; The angle between adjacent side line segments of the ceramic splitting knife is calculated through the first contour point subset and the second contour point subset to obtain the side angle of the ceramic splitting knife.
[0007] Optionally, generating grinding path information for grinding the ceramic splitting knife according to the outer contour information and the grinding wheel corner point information includes: Unifying the data formats of the outer contour information and the grinding wheel corner point information, and converting the outer contour information and the grinding wheel corner point information into a target reference coordinate system through coordinate system transformation; Determining the relative position information of the grinding wheel tool and the ceramic splitting knife according to the grinding wheel corner point information and the outer contour information; According to the outer contour information of the ceramic splitting knife and the relative position information, determine the grinding starting point and the initial grinding direction of the ceramic splitting knife, wherein the initial grinding direction is related to the outer contour tangent direction in the outer contour information; Calculate the force parameters and depth parameters for producing and grinding the ceramic splitting knife according to the size and shape of the ceramic splitting knife, the wear condition of the grinding wheel tool, and the expected grinding effect; According to the grinding starting point, the grinding initial direction, the force parameter, and the depth parameter, a grinding path for grinding on the outer contour of the ceramic splitting knife is planned, and the grinding path information for grinding the ceramic splitting knife is generated.
[0008] Optionally, planning a grinding path for grinding on the outer contour of the ceramic splitting knife according to the grinding starting point, the initial grinding direction, the force parameter, and the depth parameter, and generating the grinding path information for grinding the ceramic splitting knife includes: Determining a grinding step length according to the expected grinding effect; Taking the grinding starting point as the starting point, along the initial grinding direction, and on the outer contour of the ceramic splitting knife, generating a set of path points covering the area to be grinded on the outer contour of the ceramic splitting knife with the grinding step length; The grinding path information for grinding the ceramic splitting knife is generated according to the path point set.
[0009] Optionally, the method further includes: During the grinding process of the ceramic splitting knife, a grinding image of the ceramic splitting knife is collected to monitor the change of the outer contour of the ceramic splitting knife; By comparing the current grinding path point with the outer contour information, determining the change information of the outer contour after grinding, wherein the change information includes curve change information and angle change information; If it is detected that the outer contour changes, a new tangent direction on the outer contour is determined according to the change information, and the grinding direction of the ceramic splitting knife is dynamically adjusted based on the new tangent direction, so that the grinding wheel tool closely fits the outer contour of the ceramic splitting knife for grinding; According to the adjusted polishing direction, adjusting the polishing path after the current polishing path point; The ceramic splitting knife is continued to be ground according to the adjusted grinding path.
[0010] Optionally, the method further includes: Inputting the expected grinding effect into a target machine learning model to obtain target grinding parameters, wherein the target machine learning model is used to output the target grinding parameters according to the expected grinding effect of the ceramic splitting knife; During the grinding process of the ceramic splitting knife, a grinding image of the ceramic splitting knife is collected to monitor the change of the outer contour of the ceramic splitting knife; determining an actual polishing effect according to the polishing image; If the difference between the actual grinding effect and the expected grinding effect is greater than or equal to a preset difference threshold, adjusting the grinding path information of the ceramic splitting knife according to the target grinding parameter; The ceramic splitting knife is continued to be ground according to the adjusted grinding path.
[0011] Optionally, the method further includes: Collect parameter information of a sample ceramic splitting knife corresponding to the grinding wheel tool, as well as grinding history data of the sample ceramic splitting knife, wherein the parameter information includes the shape and size of the sample ceramic splitting knife, and the grinding history data includes the grinding parameters and grinding effect of the sample ceramic splitting knife, wherein the grinding parameters include grinding direction, grinding force, and grinding angle; Extracting a grinding effect feature set related to the grinding effect from the parameter information of the sample ceramic splitting knife and the grinding history data of the sample ceramic splitting knife; The initial machine learning model is trained by the grinding effect feature set to generate the target machine learning model, and the target machine learning model is used to output the target grinding parameters according to the expected grinding effect of the ceramic chopping knife.
[0012] In a second aspect, the present application provides an electronic device, comprising: a processor, a communication interface and a memory, wherein the processor is communicatively connected to the communication interface and the memory respectively; The memory stores computer-executable instructions; The communication interface communicates and interacts with external devices; The processor executes the computer-executable instructions stored in the memory to implement the method as described in any one of the first aspects.
[0013] In a third aspect, the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are executed by a processor, they are used to implement the method as described in any one of the first aspects.
[0014] In a fourth aspect, the present application provides a computer program product, comprising a computer program, which implements any one of the methods in the first aspect when executed by a processor.
[0015] The ceramic splitting knife size grinding method, equipment and storage medium based on image analysis provided in the present application collect the surface image of the ceramic splitting knife to be ground, analyze the surface image using an image processing model, obtain the outer contour information of the ceramic splitting knife, and obtain the grinding wheel corner point information of the grinding wheel tool for grinding the ceramic splitting knife. Based on the outer contour information and the grinding wheel corner point information, the grinding path information for grinding the ceramic splitting knife is generated, and the grinding path information includes the grinding direction, grinding force, and grinding angle. According to the grinding path information, the grinding wheel tool is controlled to grind the ceramic splitting knife. Thereby, the accuracy and efficiency of grinding the ceramic splitting knife are improved. BRIEF DESCRIPTION OF THE DRAWINGS
[0016] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.
[0017] Figure 1 A schematic diagram of a process flow of a ceramic splitting knife size grinding method based on image analysis provided in an embodiment of the present application; Figure 2 A schematic diagram of a process flow of another ceramic splitting knife size grinding method based on image analysis provided in an embodiment of the present application; Figure 3 A schematic diagram of a process flow of another ceramic splitting knife size grinding method based on image analysis provided in an embodiment of the present application; Figure 4 A schematic diagram of a process flow of another ceramic splitting knife size grinding method based on image analysis provided in an embodiment of the present application; Figure 5 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application.
[0018] The above drawings have shown clear embodiments of the present application, which will be described in more detail later. These drawings and text descriptions are not intended to limit the scope of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION
[0019] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.
[0020] The technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems are described in detail below in conjunction with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.
[0021] Figure 1 The present invention provides a flow chart of a ceramic splitting knife size grinding method based on image analysis. Figure 1 As shown, the method may include: S101, collecting a surface image of a ceramic splitting knife to be polished.
[0022] Before grinding the ceramic splitter, the server can control a high-precision industrial camera to capture images of the ceramic splitter to be ground. This process is usually carried out in a stable lighting environment to ensure the stability and consistency of image quality. The industrial camera is installed in an appropriate position to ensure that the complete surface image of the ceramic splitter can be captured. The captured image is then transmitted to the server for subsequent image processing and analysis.
[0023] For example, in an automated ceramic splitter production line, each ceramic splitter will pass through an inspection station equipped with an industrial camera before entering the grinding process. When the ceramic splitter arrives at the inspection station, the camera is automatically triggered to capture a clear surface image of the ceramic splitter. These images are then uploaded to the server for the next step of image analysis.
[0024] Among them, since the product size accuracy requirements of ceramic splitters are extremely high, such as the 1μm level tolerance requirement, the resolution requirements for industrial cameras are also extremely high. Assuming that the technical requirements of ceramic splitters are at the 0.1μm level, the length and width requirements of the ceramic splitters are 11200μm*1700μm, and the required grinding size is 1700μm*900μm, then the number of pixels required for industrial cameras is 17000*9000=153000000, that is, 150 million pixels.
[0025] S102: Analyze the surface image through an image processing model to obtain outer contour information of the ceramic splitting knife.
[0026] The outer contour information includes the bending point of the ceramic splitting knife, the head length of the ceramic splitting knife, the side length of the ceramic splitting knife, and the side angle of the ceramic splitting knife.
[0027] After receiving the surface image of the ceramic splitting knife, the server can analyze the image using a pre-trained image processing model. This model may be a deep learning model, such as a convolutional neural network (CNN), which has been trained to recognize the shape features of the ceramic splitting knife. The model first extracts the color parameters of the image and converts it into a grayscale image based on the preset color weights. The grayscale image is then normalized and denoised to improve the image quality. The processed image is input into the image processing model, which extracts the shape features of the ceramic splitting knife, including edges and corners. Based on these features, the model can calculate the head length, side length, and side angle of the ceramic splitting knife to generate complete outer contour information.
[0028] S103, obtaining the grinding wheel corner point information of the grinding wheel tool for grinding the ceramic splitter.
[0029] Before grinding, the server needs to obtain the information of the grinding wheel corner point of the grinding wheel tool. This is usually achieved by direct measurement or using a sensor. The grinding wheel corner point is an important reference point on the grinding wheel tool, which determines the position and angle of contact between the grinding wheel and the ceramic splitter. This information is crucial for generating accurate grinding paths.
[0030] For example, each grinding wheel tool is equipped with sensors to monitor the status and position of the grinding wheel in real time. When a ceramic splitter needs to be ground, the server automatically obtains the corner point information of the grinding wheel through these sensors. In this way, the server can ensure that the grinding wheel contacts the ceramic splitter at the correct angle and position.
[0031] S104: Generate grinding path information for grinding the ceramic splitting knife according to the outer contour information and the grinding wheel corner point information.
[0032] The grinding path information includes grinding direction, grinding force and grinding angle.
[0033] After obtaining the outer contour information of the ceramic splitter and the grinding wheel corner point information of the grinding wheel, the server will use this information to generate a grinding path. First, the server will unify the data formats of the two pieces of information and convert them into a common target reference coordinate system. Then, based on the grinding wheel corner point information and the outer contour information of the ceramic splitter, the server will determine the relative position of the grinding wheel tool and the ceramic splitter. Next, the server will determine the starting point and initial direction of grinding based on the outer contour information and relative position information of the ceramic splitter. Finally, based on the size and shape of the ceramic splitter, the wear of the grinding wheel, and the expected grinding effect, the server will calculate the grinding force parameters and depth parameters to plan a complete grinding path.
[0034] S105, controlling the grinding wheel tool to grind the ceramic splitting knife according to the grinding path information.
[0035] After the server generates the grinding path information, it can send this information to the robot or automated equipment that controls the grinding wheel tool. These devices will accurately control the movement trajectory, speed and force of the grinding wheel according to the received instructions to ensure that the ceramic splitting knife is ground according to the predetermined path. During the entire grinding process, the server will also monitor the grinding status and effect in real time to adjust the grinding parameters or path in time.
[0036] The method provided in the embodiment of the present application acquires the surface image of the ceramic splitting knife to be polished, analyzes the surface image using an image processing model, obtains the outer contour information of the ceramic splitting knife, and obtains the grinding wheel corner point information of the grinding wheel tool for polishing the ceramic splitting knife. Based on the outer contour information and the grinding wheel corner point information, the grinding path information for polishing the ceramic splitting knife is generated, and the grinding path information includes the grinding direction, grinding force, and grinding angle. According to the grinding path information, the grinding wheel tool is controlled to polish the ceramic splitting knife. Thereby, the accuracy and efficiency of polishing the ceramic splitting knife are improved.
[0037] In a possible implementation, the aforementioned step S102 may include: S1021. Obtain color parameters and preset color weights of the surface image.
[0038] When processing the surface image of a ceramic splitter, the server first needs to obtain the color parameters of the image. These color parameters include the red, green, and blue (RGB) values of each pixel, which represent the color information in the image. At the same time, the server also loads preset color weights, which are set based on the difference between the color characteristics of the ceramic splitter and the color of the image background. The purpose of the preset color weights is to enhance the contrast between the ceramic splitter and the background in subsequent image processing, so as to more accurately extract the shape of the ceramic splitter.
[0039] For example, in a ceramic splitting knife image analysis scenario, the server reads RGB color parameters from the image captured by the industrial camera. At the same time, based on previous experience in analyzing similar ceramic splitting knife images, a set of color weights is set. These weights may increase the difference between the unique colors of the ceramic splitting knife (such as white or light gray) and other colors in the image (such as background color or shadow color), preparing for subsequent image conversion and analysis.
[0040] S1022: Convert the surface image into a grayscale image of the surface image according to the color parameter and the preset color weight.
[0041] The server converts the original color image into a grayscale image based on the acquired color parameters and preset color weights. In this process, the server uses the color weights to adjust the RGB value of each pixel, making the ceramic splitting knife more prominent in the grayscale image. The converted grayscale image removes the color information and only retains the brightness information, simplifying the subsequent image processing steps.
[0042] For example, taking a color photo of a ceramic splitting knife, the server uses preset color weights to enhance the contrast between the ceramic splitting knife and its surroundings when converting it to a grayscale image. In this way, the outline of the ceramic splitting knife becomes clearer in the grayscale image, which facilitates the subsequent shape feature extraction.
[0043] S1023: Perform normalization and denoising processing on the grayscale image to generate a processed first image.
[0044] The server normalizes the grayscale image and adjusts the brightness value of the image to a standard range, such as between 0 and 1. This can eliminate image brightness differences caused by different lighting conditions and improve the stability of image processing. The server then performs denoising, using filters or image processing algorithms to reduce noise in the image and further highlight the outline and details of the ceramic splitting knife.
[0045] S1024: Input the first image into the pre-trained image processing model, and extract shape features of the ceramic splitting knife from the first image through the image processing model.
[0046] Among them, the image processing model includes an input layer, a convolution layer, a pooling layer, a fully connected layer, and an output layer. The convolution layer is used to extract local features in the first image, and the local features include edge features and corner features. The pooling layer is used to reduce the dimension of the local features by downsampling.
[0047] The server inputs the preprocessed first image into a pre-trained image processing model. This model may be a deep learning network, such as a convolutional neural network (CNN), which contains an input layer, a convolution layer, a pooling layer, a fully connected layer, and an output layer. The convolution layer is responsible for extracting local features in the image, such as edges and corners, while the pooling layer reduces the dimension of the features by downsampling and reduces the amount of calculation. After the model is trained, it can automatically recognize and extract the shape features of the ceramic splitting knife from the image.
[0048] On the ceramic splitter production line, the server feeds the processed first image into a trained CNN model. This model has learned how to recognize the shape features of ceramic splitters in the previous training phase. When the image passes through the convolutional layer and pooling layer of the model, the model automatically extracts key shape information such as the edges and corners of the ceramic splitter, providing a data basis for the next step of size calculation and outer contour generation.
[0049] S1025. Calculate and generate the head length, side length, and side angle of the ceramic splitting knife according to the extracted shape features of the ceramic splitting knife.
[0050] Based on the shape features extracted by the image processing model, the server will further calculate the specific size parameters of the ceramic splitting knife. By analyzing and measuring the distance between feature points, such as the length of the edge and the position of the corner point, the server can calculate the head length and side length of the ceramic splitting knife. At the same time, based on the relative position of the corner point and the direction of the edge, the server can also determine the side angle of the ceramic splitting knife.
[0051] Specifically, this step can be implemented through the following sub-steps: S1. Extracting a contour point set of the ceramic splitting knife according to shape features of the ceramic splitting knife, wherein the contour point set represents an outer boundary of the ceramic splitting knife.
[0052] After extracting the shape features of the ceramic splitting knife, the server further processes these features to extract the contour point set of the ceramic splitting knife. This contour point set is a collection of coordinate points that accurately describe the outer boundary of the ceramic splitting knife. The server may use edge detection algorithms, such as Canny edge detection, to identify and extract these contour points. These points are organized into an ordered set according to the order in which they appear in the image for subsequent analysis and processing.
[0053] S2. Matching the contour point set with the ceramic splitting knife feature region template by a shape matching algorithm to obtain the similarity between the contour point set and the ceramic splitting knife feature region template.
[0054] The ceramic splitting knife characteristic region template is a template pre-drawn according to the type and model of the ceramic splitting knife, and the ceramic splitting knife characteristic region template includes a head region and a side region.
[0055] The server uses a shape matching algorithm to compare the extracted contour point set with pre-drawn ceramic splitting knife feature area templates. These templates are pre-created according to different types of ceramic splitting knives and models, and contain key feature areas of ceramic splitting knives, such as the head area and the side area. The shape matching algorithm calculates the similarity between the contour point set and the template to determine which points belong to the head area and which points belong to the side area.
[0056] The server has a feature region template for a specific model of ceramic splitting knife. After extracting the contour point set of the ceramic splitting knife, the server will use a shape matching algorithm to compare these points with the template. This comparison process will calculate a similarity score that represents the degree of match between the contour point set and the template. This similarity score will be used to subsequently determine which points in the contour point set belong to the head area of the ceramic splitting knife and which points belong to the side area.
[0057] S3. Determine, according to the similarity, a first subset of contour points corresponding to the head region of the ceramic splitting knife and a second subset of contour points corresponding to the side region of the ceramic splitting knife.
[0058] Based on the similarity calculated by the shape matching algorithm, the server will split the contour point set into two subsets: the first contour point subset (corresponding to the head area of the ceramic splitting knife) and the second contour point subset (corresponding to the side area of the ceramic splitting knife). This segmentation process is completed based on the similarity threshold and the region boundaries defined in the template. The server ensures that the points in each subset are highly matched with their corresponding feature regions.
[0059] S4. Calculate the length of the contour line segment in the head region using the first contour point subset to obtain the head length of the ceramic splitting knife.
[0060] The server will use the first subset of contour points to calculate the length of the head area of the ceramic splitting knife. This usually involves sorting the points in the point set and calculating the sum of the distances between adjacent points to get the overall length of the head area. This process may involve some geometric calculations to ensure the accuracy of the length.
[0061] S5. Calculate the length of the contour line segment in the side area using the second contour point subset to obtain the side length of the ceramic splitting knife.
[0062] Similar to calculating the head length, the server uses the second subset of contour points to calculate the length of the side area of the ceramic riving knife. This involves sorting the points in the point set and accumulating the distances between adjacent points to get the overall length of the side area. This calculation process ensures an accurate measurement of the length of the side of the ceramic riving knife.
[0063] S6. Calculate the angle between adjacent side line segments of the ceramic splitting knife using the first contour point subset and the second contour point subset to obtain the side angle of the ceramic splitting knife.
[0064] To calculate the side angle of the ceramic splitting tool, the server uses the points in the first and second contour point subsets to determine the angle between adjacent side line segments. This typically involves calculating the angle between two line segments that are formed by the points in the contour point subsets. The server uses a geometric algorithm to accurately calculate this angle.
[0065] S1026: Generate the outer contour information according to the head length, side length and side angle of the ceramic splitting knife.
[0066] After completing all the necessary calculations and measurements, the server generates a report or data structure containing the ceramic splitting tool's outer contour information. This information not only includes the exact length of the ceramic splitting tool's head and sides, but also details the angle of the side. With this accurate outer contour information, the server can develop an efficient grinding plan to ensure that the ceramic splitting tool is ground precisely to the expected shape and size.
[0067] In a possible implementation, the aforementioned step S104 may include: S1041, unifying the data formats of the outer contour information and the grinding wheel corner point information, and converting the outer contour information and the grinding wheel corner point information into a target reference coordinate system through coordinate system transformation.
[0068] Before performing the grinding operation, the server needs to ensure that the outer contour information of the ceramic splitting tool and the grinding wheel corner point information are consistent in data format and coordinate system. First, the server will standardize the data format of these two types of information. For example, if the outer contour information is stored in the form of a series of point coordinates, and the grinding wheel corner point information is represented by a specific angle value, the server will convert this information into the same representation, such as converting them into a set of coordinate points. Next, the server will perform a coordinate system transformation. Since the outer contour information and the grinding wheel corner point information may be obtained in different coordinate systems, they need to be converted to the same target reference coordinate system through coordinate transformation. For example, linear transformation operations such as translation and rotation can be performed to ensure that the two can correctly correspond in spatial position.
[0069] For example, assume that the outer contour information received by the server is based on a local coordinate system with the center of the ceramic splitter as the origin, while the grinding wheel corner point information is based on the global coordinate system of the grinding device. The server will first identify the difference between the two coordinate systems, and then use a transformation matrix to convert the outer contour information in the local coordinate system to the global coordinate system so that it is in the same reference system as the grinding wheel corner point information. In this way, the server can accurately evaluate the relative position and posture of the grinding wheel and the ceramic splitter.
[0070] S1042, determining the relative position information of the grinding wheel tool and the ceramic splitting knife according to the grinding wheel corner point information and the outer contour information.
[0071] After unifying the data format and coordinate system, the server will use the information of the corner point of the grinding wheel and the outer contour information of the ceramic splitter to determine the relative position between them. The server will calculate the distance and direction from the corner point of the grinding wheel to the nearest point on the outer contour of the ceramic splitter, thereby obtaining the relative position relationship between the grinding wheel and the ceramic splitter.
[0072] For example, the server can calculate and find that the corner of the grinding wheel corresponds to the midpoint of one side of the ceramic splitter. By measuring the straight-line distance and angle from this midpoint to the corner of the grinding wheel, the server can accurately determine the position and posture of the grinding wheel to ensure that it can correctly contact the side of the ceramic splitter during grinding.
[0073] S1043. Determine a grinding starting point and an initial grinding direction for grinding the ceramic splitting knife according to the outer contour information of the ceramic splitting knife and the relative position information.
[0074] The initial grinding direction is related to the outer contour tangent direction in the outer contour information.
[0075] After determining the relative position of the grinding wheel and the ceramic splitter, the server needs to further determine the starting point and initial direction of grinding. The starting point is usually the point where the grinding wheel first contacts the outer contour of the ceramic splitter, which can be found by analyzing the outer contour information and relative position information. The initial grinding direction is usually consistent with the tangent direction of the outer contour at that point to ensure that the grinding wheel can grind smoothly along the outer contour of the ceramic splitter.
[0076] S1044, calculating force parameters and depth parameters for producing and grinding the ceramic splitting knife according to the size and shape of the ceramic splitting knife, the wear condition of the grinding wheel tool, and the expected grinding effect.
[0077] The server can take several factors into account when calculating the grinding force and depth parameters. The size and shape of the ceramic splitter determines the force and depth required when grinding. For example, a larger ceramic splitter may require more grinding force to ensure the grinding effect. At the same time, the wear of the grinding wheel is also an important factor. If the grinding wheel has a certain degree of wear, it may be necessary to increase the grinding depth to compensate for this wear. Finally, the desired grinding effect will also affect the choice of force and depth. If a finer grinding effect is expected, it may be necessary to reduce the grinding force and depth.
[0078] S1045. According to the grinding starting point, the grinding initial direction, the force parameter, and the depth parameter, a grinding path for grinding on the outer contour of the ceramic splitting knife is planned, and the grinding path information for grinding the ceramic splitting knife is generated.
[0079] After collecting all the necessary information, the server can start planning the grinding path. This path will specify in detail the movement trajectory of the grinding wheel on the outer contour of the ceramic splitter, including the starting point, direction, force and depth. The server can use path planning algorithms, such as interpolation or spline curve fitting, to generate a smooth and efficient grinding path. This path must not only ensure the uniformity and quality of grinding, but also take into account the efficiency of grinding and the durability of the grinding wheel.
[0080] Next, how to plan the grinding path for grinding on the outer contour of the ceramic splitting knife according to the grinding starting point, the initial grinding direction, the force parameter, and the depth parameter in the aforementioned step S1045, and generate the grinding path information for grinding the ceramic splitting knife is described in detail.
[0081] Figure 2 A schematic diagram of another method for grinding the size of a ceramic splitter based on image analysis provided in an embodiment of the present application. Figure 2 As shown, the method may include: S201. Determine a grinding step length according to the expected grinding effect.
[0082] The server can take the expected grinding effect into consideration when determining the grinding step length. The grinding step length refers to the distance covered by the grinding wheel each time it moves on the outer contour of the ceramic splitter. The setting of this parameter is crucial to the accuracy and efficiency of grinding. If the expected grinding effect is rough, a larger grinding step length can be selected to increase the grinding speed. If the expected effect is fine, a smaller grinding step length needs to be selected to ensure that each part can be finely polished. The server will automatically calculate the appropriate grinding step length based on the expected grinding effect parameters entered by the user, or based on historical data and empirical rules.
[0083] For example, suppose the user wants to finely grind a ceramic splitter to obtain a smoother surface. After receiving this instruction, the server will automatically set a smaller grinding step, such as 0.1 mm, based on historical data and empirical rules. In this way, when the grinding wheel moves along the outer contour of the ceramic splitter, only a small part of the material will be ground off each time, ensuring that the final grinding effect can meet the user's fine requirements.
[0084] S202, taking the grinding starting point as the starting point, along the initial grinding direction, and generating a set of path points covering the area to be ground on the outer contour of the ceramic splitting knife with the grinding step length.
[0085] After the server determines the grinding step length, it can gradually generate path points on the outer contour of the ceramic splitter along the initial grinding direction, starting from the grinding starting point. These path points are the positions that the grinding wheel needs to pass through in sequence during the grinding process, and the spacing between them is the previously determined grinding step length. The server can use the outer contour information of the ceramic splitter, the grinding starting point, and the initial grinding direction to accurately determine the position of each path point through mathematical calculations. This process needs to ensure that the path points can evenly and accurately cover the entire area to be ground.
[0086] S203: Generate the grinding path information for grinding the ceramic splitting knife according to the path point set.
[0087] After generating a set of path points covering the entire area to be polished, the server can generate polishing path information based on these path points. This information includes the coordinates of each path point that the grinding wheel needs to pass through in sequence, the polishing direction, force, depth and other parameters. The server will integrate this information into a complete polishing path instruction set, which will be sent to the polishing equipment to guide it to perform the polishing operation according to the predetermined path and parameters.
[0088] For example, assume that the server has generated a series of path points and calculated the corresponding parameters such as grinding direction, force and depth based on these path points. Next, the server can integrate this information into a grinding path instruction set and send it to the grinding device through the network. After receiving the instruction set, the grinding device will move to each path point in turn according to the instructions therein, and perform grinding operations according to the specified grinding direction, force and depth. In this way, the server can achieve precise control of the grinding process, thereby ensuring that the final grinding effect meets the user's requirements.
[0089] In a possible implementation, the method may further include the following contents. Figure 3 A schematic diagram of a flow chart of another method for grinding the size of a ceramic splitting knife based on image analysis provided in an embodiment of the present application. Figure 3 As shown, the method may also include: S301. In the process of grinding the ceramic splitting knife, a grinding image of the ceramic splitting knife is collected to monitor changes in the outer contour of the ceramic splitting knife.
[0090] During the grinding process of the ceramic splitter, the server controls the image acquisition device (such as a camera or sensor) to capture the grinding images of the ceramic splitter in real time. These images are collected at a certain frequency to ensure that the slight changes in the outer contour of the ceramic splitter can be captured. The collected images are transmitted to the server for further processing and analysis. In this way, the server can monitor the changes in the outer contour of the ceramic splitter during the grinding process in real time.
[0091] For example, the server is equipped with a high-definition camera, which is installed next to the grinding equipment and aimed at the ceramic splitter being ground at a fixed angle and distance. Whenever the grinding operation is carried out, the camera captures the image of the ceramic splitter at a speed of 50 frames per second and transmits the image data to the server in real time. After receiving these images, the server will perform subsequent processing and analysis.
[0092] S302, by comparing the current grinding path point with the outer contour information, determining the change information of the outer contour after grinding, wherein the change information includes curve change information and corner change information.
[0093] The server will compare the real-time acquired grinding image with the original outer contour information. Through image processing technology, such as edge detection and contour extraction, the server can identify the outer contour of the current ceramic splitter and compare it with the preset outer contour information. By calculating the difference between the two, the server can determine the changes in the outer contour of the ceramic splitter after grinding, including the degree of curvature of the curve, the size of the angle, etc.
[0094] During the grinding process, the server continuously compares the captured images with the original outer contour information. For example, the server may find that a certain side of the ceramic splitter becomes smoother after grinding, or the angle of a certain corner is reduced. These changes will be recorded by the server for subsequent grinding direction adjustments.
[0095] S303: If it is detected that the outer contour changes, a new tangent direction on the outer contour is determined according to the change information, and the grinding direction of the ceramic chopper is dynamically adjusted based on the new tangent direction, so that the grinding wheel tool closely fits the outer contour of the ceramic chopper for grinding.
[0096] Once the outer contour of the ceramic splitter is detected to have changed, the server can calculate a new tangent direction based on the change information. This new tangent direction is based on the outer contour shape of the current grinding point, and it reflects the surface direction of the ceramic splitter at that point. The server will adjust the grinding direction of the grinding wheel tool in real time to keep it consistent with the new tangent direction to ensure that the grinding wheel tool can fit the outer contour of the ceramic splitter closely for grinding.
[0097] Suppose during the grinding process, the server detects that one side of the ceramic splitter has a significant curve change. Based on this change, the server will recalculate the tangent direction of the side and immediately adjust the grinding direction of the grinding wheel tool to make it consistent with the new tangent direction. In this way, the grinding wheel tool can grind more accurately along the outer contour of the ceramic splitter, thereby improving the quality and efficiency of grinding.
[0098] S304: Adjust the polishing path after the current polishing path point according to the adjusted polishing direction.
[0099] After adjusting the grinding direction, the server also needs to adjust the subsequent grinding path accordingly. This is because the change in grinding direction may cause the original grinding path to no longer be applicable. Therefore, the server will re-plan the subsequent grinding path based on the new grinding direction and the position of the current grinding point.
[0100] Assume that during the grinding process, the server adjusts the grinding direction due to the change in the outer contour of the ceramic splitter. Then, the server will use the path planning algorithm to regenerate the subsequent grinding path based on the new grinding direction and the position of the current grinding point. In this way, the grinding wheel tool can continue the grinding operation according to the new path.
[0101] S305, continuing to grind the ceramic chopper according to the adjusted grinding path.
[0102] After adjusting the grinding path, the server will control the grinding equipment to continue grinding along the new path. During this process, the server will continue to monitor the changes in the outer contour of the ceramic splitting knife and adjust the grinding direction and path in real time as needed. In this way, the server can ensure that the grinding operation is always carried out closely in line with the outer contour of the ceramic splitting knife, so as to achieve the best grinding effect.
[0103] In a possible implementation, the method may further include the following contents. Figure 4 A schematic diagram of a flow chart of another method for grinding the size of a ceramic splitter based on image analysis provided in an embodiment of the present application. Figure 4 As shown, the method may also include: S401. Input the expected polishing effect into a target machine learning model to obtain target polishing parameters.
[0104] Among them, the target machine learning model is used to output the target grinding parameters according to the expected grinding effect of the ceramic splitting knife.
[0105] The server can receive the expected grinding effect parameters from the user interface or system settings. These parameters usually include specific requirements for the surface quality of the ceramic splitter after grinding, such as surface roughness, glossiness, flatness, etc. The server organizes these parameters into a format that the machine learning model can understand and prepares to input them into the pre-trained target machine learning model.
[0106] The target machine learning model is a complex system built on big data and deep learning algorithms. During the model training phase, a large amount of historical data and empirical knowledge related to ceramic splitting knife grinding has been used. By learning the mapping relationship between grinding parameters and grinding effects, the model can output the optimal grinding parameter combination for a specific expected grinding effect.
[0107] When the server inputs the expected grinding effect parameters into the model, the model will first parse and preprocess these parameters to ensure that they meet the model's input requirements. Then, the neural network structure inside the model will gradually extract features and learn rules based on the input parameters through multi-layer convolution, pooling, full connection and other operations, and finally output a set of optimal grinding parameters for the current ceramic splitting knife and the expected grinding effect. These parameters include but are not limited to grinding force, grinding angle, grinding speed, and possible grinding wheel type, particle size, etc.
[0108] For example, if the server receives a request for high-precision grinding of a certain model of ceramic splitting knife. The production department sets the expected grinding effect parameters, including surface roughness Ra ≤ 0.05m, gloss ≥ 90%, etc. The server formats these parameters into the input format required by the model and sends them to the target machine learning model. The model uses a complex deep learning algorithm to parse and learn the input parameters, and finally outputs a set of detailed target grinding parameters, including a grinding force of 4.8 to 5.2N, a grinding angle of 43° to 47°, and a grinding speed of 45 to 50mm / s. These parameters are verified and optimized, recorded by the server and prepared for subsequent grinding path adjustment and grinding execution.
[0109] S402: During the grinding process of the ceramic splitting knife, a grinding image of the ceramic splitting knife is collected to monitor changes in the outer contour of the ceramic splitting knife.
[0110] During the actual grinding process of the ceramic splitter, the server can control the image acquisition device (such as a high-definition camera or industrial camera) in real time to collect images of the grinding site. These image acquisition devices are usually installed near the grinding workbench to ensure that every detail change of the ceramic splitter during the grinding process can be clearly captured.
[0111] The server receives continuous image streams or image frame sequences from the image acquisition device and uses image processing technology to analyze and process these images. First, the server preprocesses the image, including denoising, contrast enhancement, brightness adjustment, etc., to improve the image quality and clarity. Then, the server uses edge detection, contour extraction and other algorithms to identify and locate the ceramic splitting knife in the image and extract its outer contour information.
[0112] After extracting the outer contour information of the ceramic splitter, the server will further analyze the changes in these contours. By comparing the contour images at different time points, the server can calculate key indicators such as edge smoothness, angle changes, and shape fine-tuning of the ceramic splitter during the grinding process. This change information is crucial for evaluating the grinding effect and adjusting the grinding path.
[0113] S403: Determine an actual polishing effect according to the polishing image.
[0114] After acquiring the polished images of the ceramic splitter and analyzing the changes in its outer contour, the server needs to further determine the actual polishing effect based on these images. The server can use image processing and analysis software to perform detailed surface quality inspections on the polished images. First, the server performs preprocessing operations such as grayscale and filtering on the images to reduce the impact of noise and interference factors on the evaluation results. Then, the server applies surface roughness measurement algorithms to evaluate the roughness of the ceramic splitter surface in the image. These algorithms estimate the microscopic geometry and roughness parameters of the surface by analyzing the changes in pixel grayscale values in the image.
[0115] In addition to roughness evaluation, the server can also use other image processing technologies to evaluate surface quality indicators such as glossiness and flatness of ceramic splitters. For example, glossiness can be evaluated by calculating the intensity distribution of reflected light on the surface of the ceramic splitter in the image; flatness can be evaluated by comparing the uniformity of image brightness in different areas.
[0116] Finally, the server will generate a detailed report on the actual polishing effect based on the evaluation results, including numerical values and charts of key indicators such as surface roughness, glossiness, flatness, etc. This report will be used for comparison and analysis with the expected polishing effect.
[0117] S404: If the difference between the actual grinding effect and the expected grinding effect is greater than or equal to a preset difference threshold, adjusting the grinding path information of the ceramic wedge according to the target grinding parameter.
[0118] After determining the actual polishing effect and comparing it with the expected effect, the server needs to evaluate the degree of difference between the two. If the difference between the actual polishing effect and the expected effect is greater than or equal to the preset difference threshold, it means that the current polishing path and parameter settings may not meet the expected polishing quality requirements and need to be adjusted and optimized.
[0119] At this point, the server will dynamically adjust the current grinding path information based on the target grinding parameters previously obtained from the target machine learning model and the current actual grinding effect feedback. This adjustment may involve multiple aspects, including fine-tuning of key parameters such as grinding force, grinding angle, and grinding speed; it may also involve re-planning the grinding path to ensure that the grinding wheel can grind the ceramic splitter along a more reasonable and effective path.
[0120] When adjusting the grinding path information, the server will take into account multiple factors, such as the shape, size, and material properties of the ceramic splitter; the type, grit, and wear of the grinding wheel; and economic factors such as production efficiency and cost. By weighing the pros and cons of these factors, the server can develop the optimal grinding path adjustment plan and implement and verify it in the actual grinding process.
[0121] S405, continuing to grind the ceramic chopper according to the adjusted grinding path.
[0122] After completing the adjustment of the grinding path, the server will control the grinding equipment to continue grinding the ceramic splitter according to the new grinding path parameters. During the grinding process, the server can continuously monitor the grinding image and actual effect to ensure that the grinding operation is carried out according to the predetermined path and parameters. If the server finds that there is still a large difference between the actual grinding effect and the expected effect or new problems have occurred (such as excessive wear of the grinding wheel, breakage of the ceramic splitter, etc.), it will immediately stop the grinding operation and conduct further inspection and analysis. After confirming the cause of the problem, the server will adjust the grinding path and parameter settings again, and restart the grinding operation until the expected effect is achieved.
[0123] In addition, during the grinding process, the server also needs to consider economic factors such as production efficiency and cost. By optimizing the grinding path and parameter settings, the server can maximize production efficiency and reduce production costs while ensuring the grinding quality. For example, by reducing unnecessary grinding paths and repeated operations to shorten the grinding time; by selecting the appropriate grinding wheel type and particle size to reduce the wear rate and replacement frequency, etc.
[0124] exist Figure 4 Under the implementation mode of , the training process of the target machine learning model is as follows: Collect parameter information of a sample ceramic splitting knife corresponding to the grinding wheel tool, as well as grinding history data of the sample ceramic splitting knife, wherein the parameter information includes the shape and size of the sample ceramic splitting knife, and the grinding history data includes grinding parameters and grinding effects of the sample ceramic splitting knife, wherein the grinding parameters include grinding direction, grinding force, and grinding angle.
[0125] In order to train a machine learning model that can accurately predict and output the target grinding parameters, the server first needs to collect a large amount of sample data. This data includes detailed parameter information and grinding history data of sample ceramic splitters corresponding to specific grinding wheel tools. The parameter information covers the physical properties of the sample ceramic splitters, such as shape and size, which are essential for understanding the structure and grinding requirements of ceramic splitters. The grinding history data records the specific parameters and final effects of the previous grinding of the sample ceramic splitters using the same or similar grinding wheel tools. These parameters include but are not limited to grinding direction, grinding force, grinding angle, etc., and the grinding effect is an evaluation of the surface quality of the ceramic splitter after grinding, such as surface roughness, glossiness, etc.
[0126] The server obtains these sample data in batches from the production record system, quality inspection system or other relevant data sources through data interfaces or database queries. During the collection process, the server will ensure the integrity, accuracy and consistency of the data, and perform necessary cleaning and preprocessing on missing or abnormal data to ensure the effectiveness and reliability of subsequent model training.
[0127] A grinding effect feature set related to the grinding effect is extracted from the parameter information of the sample ceramic cleaver and the grinding history data of the sample ceramic cleaver.
[0128] After collecting enough sample data, the server needs to extract a set of features that are closely related to the grinding effect. These features should be able to fully and accurately reflect the physical properties of the ceramic splitter, the key parameters in the grinding process, and the final grinding effect. The server will use feature engineering technology to screen out the feature variables that have the greatest impact on the grinding effect through statistical analysis, correlation analysis, principal component analysis and other methods. These features may include specific size ratios of ceramic splitters, combinations of grinding parameters, and their interactions with grinding effects.
[0129] In the process of extracting feature sets, the server also needs to consider the interpretability of features and the generalization ability of the model. Features with strong interpretability help us understand the working principle and decision basis of the model; while models with strong generalization ability can maintain high prediction accuracy when facing new samples. Therefore, the server will weigh these two factors when selecting features, striving to improve the practical application value of the model while maintaining its prediction accuracy.
[0130] The initial machine learning model is trained by the grinding effect feature set to generate the target machine learning model, and the target machine learning model is used to output the target grinding parameters according to the expected grinding effect of the ceramic chopping knife.
[0131] After extracting the feature set related to the polishing effect, the server will use these features to train the initial machine learning model. The training process usually includes steps such as data partitioning, model selection, parameter tuning, and performance evaluation. First, the server will divide the sample data into two parts: a training set and a test set. The training set is used to train the model parameters, and the test set is used to evaluate the model performance. Then, the server will select a suitable machine learning algorithm to build the initial model based on the nature of the problem and the characteristics of the data. During the model training process, the server will continuously adjust the parameters of the model to optimize its performance on the training set. Finally, the server will use the test set to evaluate the generalization ability of the trained model, and further tune the model or select other algorithms for retraining based on the evaluation results.
[0132] During the training process, the server also uses cross-validation and other techniques to improve the stability and reliability of the model. Cross-validation further divides the training set into multiple subsets and uses them as validation sets in turn to evaluate model performance, thereby avoiding overfitting. After multiple iterations of training and parameter adjustment, the server will eventually generate a target machine learning model with excellent performance for practical applications.
[0133] Figure 5 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. The electronic device is used to execute the ceramic splitting knife size grinding method based on image analysis, for example, it can be the server mentioned above. Figure 5 As shown, the electronic device 500 may include: at least one processor 501 , a memory 502 , and a communication interface 503 .
[0134] The memory 502 is used to store programs. Specifically, the programs may include program codes, and the program codes include computer operation instructions.
[0135] The memory 502 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.
[0136] The processor 501 is used to execute the computer-executable instructions stored in the memory 502 to implement the method described in the above method embodiment. The processor 501 may be a CPU, or an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application.
[0137] The processor 501 can communicate and interact with external devices through the communication interface 503. In specific implementation, if the communication interface 503, the memory 502 and the processor 501 are implemented independently, the communication interface 503, the memory 502 and the processor 501 can be connected to each other through a bus and communicate with each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc., but it does not mean that there is only one bus or one type of bus.
[0138] Optionally, in a specific implementation, if the communication interface 503, the memory 502 and the processor 501 are integrated on a chip, the communication interface 503, the memory 502 and the processor 501 can complete communication through an internal interface.
[0139] The present application also provides a computer-readable storage medium, which may include: a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk, and other media that can store program codes. Specifically, the computer-readable storage medium stores program instructions, and the program instructions are used for the methods in the above embodiments.
[0140] The present application also provides a program product, which includes an execution instruction, which is stored in a readable storage medium. At least one processor of a computing device can read the execution instruction from the readable storage medium, and at least one processor executes the execution instruction so that the computing device implements the above method.
[0141] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.
Claims
1. A ceramic splitter size grinding method based on image analysis, characterized in that: The method comprises: Acquiring a surface image of a ceramic splitting knife to be sharpened; Analyzing the surface image through an image processing model to obtain the outer contour information of the ceramic splitting knife, wherein the outer contour information includes the bending point of the ceramic splitting knife, the head length of the ceramic splitting knife, the side length of the ceramic splitting knife, and the side angle of the ceramic splitting knife; Acquire the grinding wheel corner point information of the grinding wheel tool for grinding the ceramic splitter; Generate grinding path information for grinding the ceramic splitting knife according to the outer contour information and the grinding wheel corner point information, wherein the grinding path information includes a grinding direction, a grinding force, and a grinding angle; Controlling the grinding wheel tool to grind the ceramic splitting knife according to the grinding path information; The step of analyzing the surface image by an image processing model to obtain the outer contour information of the ceramic splitting knife includes: Obtaining color parameters and preset color weights of the surface image; converting the surface image into a grayscale image of the surface image according to the color parameter and the preset color weight; Performing normalization and denoising processing on the grayscale image to generate a processed first image; Inputting the first image into the pre-trained image processing model, and extracting the shape features of the ceramic splitting knife from the first image through the image processing model; the image processing model includes an input layer, a convolution layer, a pooling layer, a fully connected layer, and an output layer, the convolution layer is used to extract local features in the first image, the local features include edge features and corner features, and the pooling layer is used to reduce the dimension of the local features by downsampling; According to the extracted shape features of the ceramic splitting knife, the head length, side length and side angle of the ceramic splitting knife are calculated and generated; Generate the outer contour information according to the head length, side length and side angle of the ceramic splitting knife; The step of calculating and generating the head length, side length and side angle of the ceramic splitting knife according to the extracted shape features of the ceramic splitting knife comprises: Extracting a contour point set of the ceramic splitting knife according to the shape feature of the ceramic splitting knife, wherein the contour point set represents the outer boundary of the ceramic splitting knife; Matching the contour point set and the ceramic splitting knife feature region template by a shape matching algorithm to obtain the similarity between the contour point set and the ceramic splitting knife feature region template, wherein the ceramic splitting knife feature region template is a template pre-drawn according to the type and model of the ceramic splitting knife, and the ceramic splitting knife feature region template includes a head region and a side region; According to the similarity, determining a first subset of contour points corresponding to the head region of the ceramic splitting knife and a second subset of contour points corresponding to the side region of the ceramic splitting knife; Calculating the length of the contour line segment in the head region by using the first contour point subset to obtain the head length of the ceramic splitting knife; Calculating the length of the contour line segment in the side area by using the second contour point subset to obtain the side length of the ceramic splitting knife; Calculating the angle between adjacent side line segments of the ceramic splitting knife through the first contour point subset and the second contour point subset to obtain the side angle of the ceramic splitting knife; The step of generating grinding path information for grinding the ceramic splitting knife according to the outer contour information and the grinding wheel corner point information includes: Unifying the data formats of the outer contour information and the grinding wheel corner point information, and converting the outer contour information and the grinding wheel corner point information into a target reference coordinate system through coordinate system transformation; Determining the relative position information of the grinding wheel tool and the ceramic splitting knife according to the grinding wheel corner point information and the outer contour information; According to the outer contour information of the ceramic splitting knife and the relative position information, determine the grinding starting point and the initial grinding direction of the ceramic splitting knife, wherein the initial grinding direction is related to the outer contour tangent direction in the outer contour information; Calculate the force parameters and depth parameters for producing and grinding the ceramic splitting knife according to the size and shape of the ceramic splitting knife, the wear condition of the grinding wheel tool, and the expected grinding effect; According to the grinding starting point, the grinding initial direction, the force parameter, and the depth parameter, a grinding path for grinding on the outer contour of the ceramic splitting knife is planned, and the grinding path information for grinding the ceramic splitting knife is generated; The step of planning a grinding path for grinding on the outer contour of the ceramic splitting knife according to the grinding starting point, the grinding initial direction, the force parameter, and the depth parameter, and generating the grinding path information for grinding the ceramic splitting knife includes: Determining a grinding step length according to the expected grinding effect; Taking the grinding starting point as the starting point, along the initial grinding direction, and on the outer contour of the ceramic splitting knife, generating a set of path points covering the area to be grinded on the outer contour of the ceramic splitting knife with the grinding step length; The grinding path information for grinding the ceramic splitting knife is generated according to the path point set.
2. The method according to claim 1, characterized in that The method further comprises: During the grinding process of the ceramic splitting knife, a grinding image of the ceramic splitting knife is collected to monitor the change of the outer contour of the ceramic splitting knife; By comparing the current grinding path point with the outer contour information, determining the change information of the outer contour after grinding, wherein the change information includes curve change information and angle change information; If it is detected that the outer contour changes, a new tangent direction on the outer contour is determined according to the change information, and the grinding direction of the ceramic splitting knife is dynamically adjusted based on the new tangent direction, so that the grinding wheel tool closely fits the outer contour of the ceramic splitting knife for grinding; According to the adjusted polishing direction, adjusting the polishing path after the current polishing path point; The ceramic splitting knife is continued to be ground according to the adjusted grinding path.
3. The method according to claim 1 or 2, characterized in that: The method further comprises: Inputting the expected grinding effect into a target machine learning model to obtain target grinding parameters, wherein the target machine learning model is used to output the target grinding parameters according to the expected grinding effect of the ceramic splitting knife; During the grinding process of the ceramic splitting knife, a grinding image of the ceramic splitting knife is collected to monitor the change of the outer contour of the ceramic splitting knife; determining an actual polishing effect according to the polishing image; If the difference between the actual grinding effect and the expected grinding effect is greater than or equal to a preset difference threshold, adjusting the grinding path information of the ceramic splitting knife according to the target grinding parameter; The ceramic splitting knife is continued to be ground according to the adjusted grinding path.
4. The method according to claim 3, characterized in that The method further comprises: Collect parameter information of a sample ceramic splitting knife corresponding to the grinding wheel tool, as well as grinding history data of the sample ceramic splitting knife, wherein the parameter information includes the shape and size of the sample ceramic splitting knife, and the grinding history data includes the grinding parameters and grinding effect of the sample ceramic splitting knife, wherein the grinding parameters include grinding direction, grinding force, and grinding angle; Extracting a grinding effect feature set related to the grinding effect from the parameter information of the sample ceramic splitting knife and the grinding history data of the sample ceramic splitting knife; The initial machine learning model is trained by the grinding effect feature set to generate the target machine learning model, and the target machine learning model is used to output the target grinding parameters according to the expected grinding effect of the ceramic chopping knife.
5. An electronic device, characterized in that: include: A processor, a communication interface and a memory, wherein the processor is communicatively connected to the communication interface and the memory respectively; The memory stores computer-executable instructions; The communication interface communicates and interacts with external devices; The processor executes the computer-executable instructions stored in the memory to implement the method according to any one of claims 1 to 4.
6. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 4 when executed by a processor.
7. A computer program product, characterized in that The invention comprises a computer program, which implements the method according to any one of claims 1 to 4 when being executed by a processor.
Citation Information
Patent Citations
Grinding equipment for grinding inner chamfer of ceramic chopper
CN114750066A
Grinding mechanism for grinding inner chamfer of ceramic chopper
CN114750067A
Grinding mechanism for grinding outer diameter of ceramic chopper
CN114770304A
Ceramic chopper outer radius grinding method, system, equipment and medium
CN115781424A
Chopping knife and grinding process
CN117984164A
Cited By
Ceramic chopper automatic concentricity adjustment grinding technology based on machine vision
CN119141338A
A method for automatically adjusting the concentricity of ceramic splitting cutters based on machine vision
CN119141338B
