Height correction method and system based on machine vision measurement

Abstract

The invention relates to a height correction method and system based on machine vision measurement, and the method comprises the steps: obtaining a film mark image on a wafer, and measuring the length of a tool mark of a cutting tool from the film mark image through a machine vision technology; calculating the actual cutting depth of the cutting tool according to the length of the tool mark; the difference value between the preset depth value and the actual cut-in depth is calculated, the height correction value of the cutting tool is calculated and determined, the height correction value is dynamically adjusted according to the attribute of the cutting tool, and the attribute of the cutting tool comprises the material and the working state; the height of the cutting tool is adjusted according to the height correction value, the moving path of the cutting tool is monitored in real time in the adjusting process of the cutting tool, and if it is monitored that the deviation of the moving path exceeds a preset threshold value, a path correction prompt is sent out. The cutting device has the effect of improving the cutting precision.

Description

Technical Field

[0001] The present application relates to the technical field of semiconductor wafer cutting, and in particular to a height correction method and system based on machine vision measurement. Background Art

[0002] Currently, in the semiconductor manufacturing process, the wafer cutting accuracy directly affects the quality and yield of the chip. Usually, the height of the cutting tool needs to be precisely adjusted to ensure that the cutting depth meets the requirements. However, traditional cutting tool height adjustment methods mostly rely on manual experience or preset depth values. These methods are not only complex and error-prone, but also cannot adapt to changes in different materials and cutting conditions.

[0003] Existing technical solutions usually rely on image detection or film mark measurement to assist in adjusting the tool height. In these methods, the film mark length is used to estimate the tool's cutting depth, which is then corrected by adjusting the tool height. However, due to factors such as illumination, reflection, and surface unevenness on the wafer surface, traditional image detection methods often have difficulty providing high-precision measurement results. In addition, tool wear, temperature changes, and other environmental factors may also affect the cutting depth, but traditional methods cannot make real-time adaptive adjustments to these factors, resulting in instability and errors in the cutting process.

[0004] The above-mentioned existing technical solutions have the following defects: the existing tool height correction method often relies on a static standard depth value, fails to consider the impact of factors such as the tool material and working status on the cutting depth, and cannot perform real-time dynamic adjustments according to actual conditions, resulting in limited cutting accuracy. Therefore, there is room for improvement. Summary of the invention

[0005] In order to improve cutting accuracy, the present application provides a height correction method and system based on machine vision measurement.

[0006] The above-mentioned invention objective of the present application is achieved through the following technical solutions: A height correction method based on machine vision measurement, the method comprising: obtaining cutting parameter data, obtaining cutting direction data and cutting sequence data corresponding to each cutting direction data from the cutting parameter data, and generating a wafer cutting instruction according to the cutting sequence data; Acquire a film mark image on the wafer, and use machine vision technology to measure the length of the cutter mark of the cutting tool from the film mark image; Calculating the actual cutting depth of the cutting tool according to the length of the cutter mark; Calculating the difference between the preset depth value and the actual cutting depth, calculating and determining the height correction value of the cutting tool, and dynamically adjusting the height correction value according to the properties of the cutting tool, wherein the properties of the cutting tool include material and working status; The height of the cutting tool is adjusted according to the height correction value, and the moving path of the cutting tool is monitored in real time during the adjustment process of the cutting tool. If it is monitored that the moving path deviation exceeds a preset threshold, a path correction prompt is issued.

[0007] By adopting the above technical solution, by obtaining cutting parameter data and extracting cutting direction data and cutting sequence data therefrom, the cutting path and sequence can be accurately determined, so that the cutting task can be executed according to the optimized path sequence, thereby improving cutting efficiency and accuracy; by using machine vision technology to obtain the film mark image on the wafer and measure the knife mark length of the cutting tool, the cutting depth can be monitored in real time, thereby ensuring the accurate cutting depth of the tool during the cutting process; by calculating the difference between the preset depth value and the actual cutting depth and determining the height correction value, the height of the cutting tool can be accurately adjusted, thereby effectively avoiding cutting errors; by dynamically adjusting the height correction value according to the properties of the tool, it can adapt to different tools and working environments, and improve the accuracy and flexibility of height adjustment; by real-time monitoring the movement path of the tool and correcting the path deviation, it can ensure the stability of the cutting path, avoid cutting errors caused by path deviations, and thus improve the cutting quality.

[0008] In one example, the present application may be further configured as follows: the obtaining of cutting direction data and cutting sequence data corresponding to each cutting direction data from the cutting parameter data, and generating a wafer cutting instruction according to the cutting sequence data includes: Extracting a cutting area of ​​the wafer from the cutting parameter data and identifying a cutting direction of the wafer; Based on the cutting direction, determine the starting point and the ending point of each cutting direction, and according to the data of the cutting direction, determine the cutting order of each direction to obtain the cutting order data; In combination with the cutting sequence data, the corresponding wafer cutting instruction is generated to control the cutting tool to perform the cutting task according to a predetermined path sequence.

[0009] By adopting the above technical solution, the cutting area is determined and the cutting direction is identified through the basic cutting configuration data in the cutting parameters, so that accurate path data can be provided for the cutting task, thereby ensuring that the order and position of each cutting direction are accurate; determining the starting point and end point of each cutting direction, and generating the cutting sequence based on these data, can achieve efficient planning of the cutting task and ensure that the cutting process proceeds smoothly according to the predetermined path; combining the cutting sequence data to generate wafer cutting instructions can enable the cutting tool to perform tasks according to the precise path sequence, further improving the automation and efficiency of cutting.

[0010] In one example, the present application can be further configured as follows: the obtaining of the film mark image of the cutting tool on the wafer includes: photographing the cutting marks on the wafer using microscopy technology, and extracting the cutting edges of the cutting marks using an image segmentation algorithm to generate the film mark image.

[0011] By adopting the above technical solution, by using microscope technology to photograph the cutting marks on the wafer and using image segmentation algorithm to extract the cutting edge, the position of the knife mark in the film mark image can be accurately identified, ensuring the high resolution of the image and accurate measurement of the cutting depth; by extracting the cutting edge through the image segmentation algorithm, the accuracy of film mark image processing can be greatly improved, thereby providing accurate data for subsequent cutting depth measurement and tool adjustment.

[0012] In one example, the present application may be further configured as follows: the step of measuring the length of the cutting tool mark from the film mark image using machine vision technology includes: Generate a first camera movement instruction, and obtain the first cut mark endpoint coordinates according to the first camera movement instruction; When the first cut endpoint information is obtained, a camera return instruction is generated; When receiving the camera return instruction, generating a second camera movement instruction, and acquiring the second cut mark endpoint coordinates according to the second camera movement instruction; The cut mark length is calculated based on the first cut mark endpoint coordinates and the second cut mark endpoint coordinates.

[0013] By adopting the above technical solution, by generating a first camera movement instruction and obtaining the coordinates of the first cut mark endpoint, the starting point of the tool mark can be accurately measured to ensure the position accuracy during the measurement process; by generating a second camera movement instruction and obtaining the coordinates of the second cut mark endpoint, the complete tool mark length can be obtained to ensure the accuracy of the cutting depth measurement; by calculating the distance between the first and second cut mark endpoint coordinates, the actual length of the tool mark can be obtained, providing accurate data for height correction, thereby ensuring the accuracy of the tool height adjustment.

[0014] In one example, the present application may be further configured as follows: calculating the actual cutting depth of the cutting tool according to the cut length includes: According to the Pythagorean theorem: a 2 +b 2 =c 2 , calculate the distance from the cutting tool to the wafer surface, where c is the radius of the cutting tool, 2a is the length of the cutter mark, and b is the distance from the cutting tool to the wafer surface; The difference between the radius of the cutting tool and the distance from the cutting tool to the wafer surface is calculated to obtain the actual cutting depth of the cutting tool.

[0015] By adopting the above technical solution, by calculating the distance from the cutting tool to the wafer surface according to the Pythagorean theorem, the actual cutting depth of the cutting tool can be accurately calculated according to the length of the knife mark, thereby avoiding cutting problems caused by depth calculation errors; by calculating the difference between the set depth and the actual cutting depth, an accurate tool height correction value can be obtained, thereby ensuring that the cutting tool accurately cuts into the wafer surface and avoiding the influence of depth error on cutting quality.

[0016] In one example, the present application may be further configured as follows: the height of the cutting tool is adjusted according to the height correction value, and the moving path of the cutting tool is monitored in real time during the adjustment of the cutting tool. If the moving path deviation is detected to exceed a preset threshold, a path correction prompt is issued, including: According to the height correction value, controlling the tool adjustment system to adjust the cutting tool; During the adjustment process of the cutting tool, the moving path of the cutting tool is monitored by a motor sensor, and the deviation between the moving path of the cutting tool and a preset path is compared to obtain the moving path deviation; When the moving path deviation exceeds the preset threshold, the path correction prompt is triggered.

[0017] By adopting the above technical solution, by adjusting the height of the cutting tool according to the height correction value, the height error of the tool can be corrected in real time to ensure that the contact depth between the tool and the wafer surface during the cutting process meets the predetermined standard; by real-time monitoring the movement path of the tool and comparing the deviation between the path and the preset path, the path deviation can be discovered and corrected in time, thereby avoiding the path error affecting the cutting accuracy; when the path deviation exceeds the preset threshold, the path correction prompt is triggered to ensure the stability of the cutting path and the cutting quality.

[0018] In one example, the present application may be further configured as follows: the height correction method based on machine vision measurement also includes: Before each cutting, the height of the cutting tool is adjusted in combination with the sensor feedback signal; During the cutting process, an adaptive algorithm is used to calibrate the height of the cutting tool in real time in combination with the flatness of the cutting table where the wafer is located; When the wafer first direction cutting completion message is obtained, a cutting table rotation instruction is generated; When the message indicating that the cutting table rotation is completed is obtained, a second direction wafer position image is obtained.

[0019] By adopting the above technical solution, the height of the cutting tool is adjusted in combination with the sensor feedback signal, so that the height of the tool can be adjusted in real time to ensure that each cutting meets the precision requirements; by adopting an adaptive algorithm combined with the flatness of the wafer cutting table for real-time calibration, the height error caused by the unevenness of the cutting table can be automatically compensated, thereby improving the stability and accuracy of the cutting.

[0020] The second object of the invention is achieved by the following technical solutions: A height correction system based on machine vision measurement, the height correction system based on machine vision measurement comprising: a cutting module, used to obtain cutting parameter data, obtain cutting direction data and cutting sequence data corresponding to each cutting direction data from the cutting parameter data, and generate a wafer cutting instruction according to the cutting sequence data; An image acquisition module is used to acquire a film mark image on the wafer, and to measure the length of a cutter mark of a cutting tool from the film mark image using machine vision technology; A module for calculating actual depth, used for calculating the actual cutting depth of the cutting tool according to the length of the cutter mark; A correction value calculation module, used for calculating the difference between a preset depth value and the actual cutting depth, calculating and determining a height correction value of the cutting tool, and dynamically adjusting the height correction value according to the properties of the cutting tool, wherein the properties of the cutting tool include material and working status; The adjustment module is used to adjust the height of the cutting tool according to the height correction value, and monitor the moving path of the cutting tool in real time during the adjustment process of the cutting tool. If the moving path deviation is detected to exceed a preset threshold, a path correction prompt is issued.

[0021] By adopting the above technical solutions, by obtaining the cutting parameter data and extracting the cutting direction data and cutting sequence data therefrom, the cutting path and sequence can be accurately determined, so that the cutting task can be executed according to the optimized path sequence, improving the cutting efficiency and accuracy; by using machine vision technology to obtain the film mark image on the wafer and measure the length of the tool mark of the cutting tool, the cutting depth can be monitored in real time, thus ensuring the accurate cutting depth of the tool during the cutting process; by calculating the difference between the preset depth value and the actual cutting depth and determining the height correction value, the height of the cutting tool can be accurately adjusted, thus effectively avoiding cutting errors; by dynamically adjusting the height correction value according to the attributes of the tool, different tools and working environments can be adapted, improving the accuracy and flexibility of height adjustment; by monitoring the moving path of the tool in real time and correcting the path deviation, the cutting path can be ensured to be stable, avoiding cutting errors caused by path deviation, and thus improving the cutting quality.

[0022] In summary, the present application includes the following beneficial technical effects: 1. The cutting path and sequence can be accurately determined, so that the cutting task can be executed according to the optimized path sequence, improving the cutting efficiency and accuracy; by using machine vision technology to obtain the film mark image on the wafer and measure the length of the tool mark of the cutting tool, the cutting depth can be monitored in real time, thus ensuring the accurate cutting depth of the tool during the cutting process; 2. By calculating the difference between the preset depth value and the actual cutting depth and determining the height correction value, the height of the cutting tool can be accurately adjusted, thus effectively avoiding cutting errors; by dynamically adjusting the height correction value according to the attributes of the tool, different tools and working environments can be adapted, improving the accuracy and flexibility of height adjustment; by monitoring the moving path of the tool in real time and correcting the path deviation, the cutting path can be ensured to be stable, avoiding cutting errors caused by path deviation, and thus improving the cutting quality. Description of the Drawings

[0023] Figure 1 is a flowchart of a height correction method based on machine vision measurement in an embodiment of the present application; Figure 2 is an implementation flowchart of step S10 in a height correction method based on machine vision measurement in an embodiment of the present application; Figure 3 is an implementation flowchart of step S20 in a height correction method based on machine vision measurement in an embodiment of the present application; Figure 4 is an implementation flowchart of step S20 in a height correction method based on machine vision measurement in an embodiment of the present application; Figure 5is a flowchart for implementing step S30 in a height correction method based on machine vision measurement in one embodiment of the present application; Figure 6 is a flowchart for implementing step S40 in a height correction method based on machine vision measurement in one embodiment of the present application; Figure 7 This is a flowchart of an implementation of a height correction method based on machine vision measurement in one embodiment of the present application; Figure 8 It is a principle block diagram of a height correction system based on machine vision measurement in one embodiment of the present application. DETAILED DESCRIPTION

[0024] The present application is further described in detail below in conjunction with the accompanying drawings.

[0025] In one embodiment, if Figure 1 As shown, the present application discloses a height correction method based on machine vision measurement, which specifically includes the following steps: S10: Obtain cutting parameter data, obtain cutting direction data and cutting sequence data corresponding to each cutting direction data from the cutting parameter data, and generate a wafer cutting instruction according to the cutting sequence data.

[0026] Specifically, first, all cutting-related data, including cutting path, direction, sequence, etc., are obtained through a preset database or parameter file, and these data are processed using an analytical algorithm to extract each cutting direction and the corresponding cutting sequence to ensure the accuracy and completeness of the data. Then, corresponding cutting instructions are generated based on these data. These instructions will be used to control each cutting step in the cutting process to ensure that the cutting task is completed in the correct order and path, and ultimately achieve precise cutting of the wafer. In this embodiment, the cutting direction data includes a first cutting direction and a second cutting direction, and the cutting sequence is to execute the first cutting direction first and then the second cutting direction. In other embodiments, different cutting directions and cutting sequences can be set according to the specific shape of the wafer after cutting.

[0027] S20: Acquire a film mark image on the wafer, and use machine vision technology to measure the length of the cutting tool mark from the film mark image.

[0028] Specifically, the film marks generated during the wafer surface cutting process are photographed using microscope technology and a high-definition camera to ensure that the obtained film mark images have sufficient details and resolution. Then, image segmentation is performed through image processing software to automatically extract the contour edges of the film marks, and advanced image analysis algorithms such as edge detection or region growth algorithms are used to accurately identify the actual position of the film marks. Next, by calculating the actual length of the film mark and combining it with relevant parameters such as the radius of the tool and the cutting angle, the actual cutting depth of the tool on the wafer surface can be accurately obtained.

[0029] S30: Calculate the actual cutting depth of the cutting tool according to the length of the cutter mark.

[0030] Specifically, based on the measured film mark length, the radius of the cutting tool and the cutting angle, the actual cutting depth of the tool from the wafer surface is calculated by a geometric calculation formula or a three-dimensional modeling method. For example, a mathematical model based on the tool radius and film mark length is used to calculate the actual contact depth between the tool and the wafer surface in combination with geometric principles.

[0031] S40: Calculate the difference between the preset depth value and the actual cutting depth, calculate and determine the height correction value of the cutting tool, and dynamically adjust the height correction value according to the properties of the cutting tool. The properties of the cutting tool include material and working status.

[0032] Specifically, the difference between the set standard preset depth value and the actual cutting depth calculated previously is calculated to obtain a preliminary tool height correction value. This correction value is used to adjust the height of the tool so that it can accurately reach the predetermined cutting depth. However, since the cutting tool will be subject to wear, temperature changes and other factors during use, which affect the cutting ability of the tool, it is necessary to make dynamic adjustments based on the specific properties of the tool. For example, the material of the tool (such as hardness, coating, etc.) and the current working state (such as temperature, degree of wear) will affect its cutting accuracy. By real-time monitoring of the working state of the tool (for example, through a temperature sensor or a pressure sensor), the height correction value is adjusted in real time in combination with these properties to ensure that the tool always maintains the best working state, thereby improving cutting accuracy and stability.

[0033] S50: adjusting the height of the cutting tool according to the height correction value, and monitoring the moving path of the cutting tool in real time during the adjustment process of the cutting tool, and issuing a path correction prompt if it is detected that the moving path deviation exceeds a preset threshold.

[0034] Specifically, the height of the tool is adjusted according to the calculated height correction value so that it can accurately reach the target cutting depth. During this process, the tool's moving path is monitored in real time, and the tool's displacement and trajectory are continuously tracked using position sensors or machine vision technology to ensure that the tool always stays on the correct path during the cutting process. By comparing with the preset path in real time, if the tool's moving path deviates and the deviation exceeds the preset threshold, the system will automatically trigger the path correction mechanism, prompting the operator or automatically adjusting the tool path to ensure that the tool's path correction during the cutting process is timely and accurate, thereby avoiding the adverse effects of path deviation on cutting accuracy and improving the overall cutting quality.

[0035] By adopting the above technical solution, by obtaining cutting parameter data and extracting cutting direction data and cutting sequence data therefrom, the cutting path and sequence can be accurately determined, so that the cutting task can be executed according to the optimized path sequence, thereby improving cutting efficiency and accuracy; by using machine vision technology to obtain the film mark image on the wafer and measure the knife mark length of the cutting tool, the cutting depth can be monitored in real time, thereby ensuring the accurate cutting depth of the tool during the cutting process; by calculating the difference between the preset depth value and the actual cutting depth and determining the height correction value, the height of the cutting tool can be accurately adjusted, thereby effectively avoiding cutting errors; by dynamically adjusting the height correction value according to the properties of the tool, it can adapt to different tools and working environments, and improve the accuracy and flexibility of height adjustment; by real-time monitoring the movement path of the tool and correcting the path deviation, it can ensure the stability of the cutting path, avoid cutting errors caused by path deviations, and thus improve the cutting quality.

[0036] In one embodiment, if Figure 2 As shown, in step S10, the cutting direction data and the cutting sequence data corresponding to each cutting direction data are obtained from the cutting parameter data, and a wafer cutting instruction is generated according to the cutting sequence data, specifically including: S11: extracting the cutting area of ​​the wafer from the cutting parameter data and identifying the cutting direction of the wafer.

[0037] Specifically, by reading the area setting information in the cutting parameter data, the wafer image is divided into multiple areas, and each specific cutting area is extracted according to the cutting area coordinates or the set range in the data. Then, the cutting direction of the wafer is determined according to the cutting direction data in the parameters. The cutting direction of each area can be determined by analyzing the angle information in the data or by relative coordinate positioning to ensure that the cutting direction is consistent with the actual geometric shape and requirements of the wafer, and finally obtain accurate cutting direction data.

[0038] S12: Based on the cutting direction, determine the starting point and the ending point of each cutting direction, and determine the cutting order of each direction according to the cutting direction data to obtain cutting order data.

[0039] Specifically, through the cutting direction data identified in the previous step, the morphological algorithm or contour analysis algorithm is applied to further identify the precise starting point and end point of each cutting direction, and the path length and position of each cutting direction are determined in combination with geometric analysis. Then, according to the size, material properties and cutting rules of the wafer, the cutting sequence data of each cutting direction is generated through the optimization algorithm to ensure that the cutting operation is performed in the best order, thereby improving the cutting efficiency and accuracy.

[0040] S13: Generate corresponding wafer cutting instructions in combination with the cutting sequence data to control the cutting tool to perform the cutting task according to the predetermined path sequence.

[0041] Specifically, by combining the cutting sequence data with the path control data, precise cutting instructions are generated based on the path planning algorithm. These instructions are converted into control signals and sent to the cutting equipment to ensure that the cutting tool performs the cutting task in accordance with the planned path sequence, thereby optimizing the cutting process, avoiding path intersection and unnecessary movement, and improving the cutting automation level and work efficiency.

[0042] In one embodiment, if Figure 3 As shown, in step S20, the film mark image of the cutting tool on the wafer is obtained, which specifically includes: S21: Use microscope technology to photograph the cutting marks on the wafer, and use image segmentation algorithm to extract the cutting edges of the cutting marks to generate a film mark image.

[0043] Specifically, a high-precision microscope or microscope camera is used to photograph the wafer surface to obtain a clear image of the cutting mark. Image processing techniques, such as edge detection and binarization, are applied to extract the edge of the cutting mark from the background and generate a film mark image for further analysis and calculation to ensure the accuracy and clarity of the cutting mark.

[0044] In one embodiment, if Figure 4 As shown, in step S20, the length of the cutter mark of the cutting tool is measured from the film mark image using machine vision technology, which specifically includes: S22: Generate a first camera movement instruction, and obtain the first cut end point coordinates according to the first camera movement instruction.

[0045] Specifically, the center point coordinates of the wafer are first located, and a coordinate system is constructed based on the center point coordinates. In the constructed coordinate system, the camera movement instructions are generated according to the starting point of the cutting mark, so that the camera moves along the cutting path from the starting point until it approaches the endpoint of the first cutting mark, and the coordinate data of the endpoint is obtained through a precise positioning algorithm.

[0046] S23: When the first cut endpoint information is obtained, a camera return instruction is generated.

[0047] Specifically, after the high-precision microscope camera obtains the coordinates of the endpoint of the first cut, a camera return instruction is generated through the control algorithm, instructing the camera to return to the starting point in the opposite direction to prepare for the subsequent acquisition of the coordinates of the second cut, ensuring the continuity of the operation and the accuracy of the position.

[0048] S24: When receiving the camera return instruction, a second camera movement instruction is generated, and the coordinates of the second cut end point are obtained according to the second camera movement instruction.

[0049] Specifically, after the camera returns to the starting point and completes positioning, a second camera movement instruction is generated to cause the camera to continue moving along the cutting path until the endpoint coordinates of the second cut are obtained, ensuring that the camera can accurately reach the end point of the second cut, thereby obtaining complete cutting mark data.

[0050] S25: Calculate the length of the cut mark according to the coordinates of the first cut mark endpoint and the coordinates of the second cut mark endpoint.

[0051] Specifically, accurate geometric data is provided for subsequent depth calculations to ensure the accuracy of tool mark length measurement. The length of the tool mark is obtained by calculating the straight-line distance between the first cut end point and the second cut end point. In the present embodiment, when obtaining the endpoint coordinates of the first cut end point and the second cut end point, the first cut end point coordinate information and the second cut end point coordinate information of multiple different positions are obtained, and the corresponding first cut end point coordinate information and the second cut end point coordinate information are calculated. Then, the calculated values ​​are averaged to finally calculate the tool mark length of the cutting tool.

[0052] In one embodiment, if Figure 5 As shown, in step S30, the actual cutting depth of the cutting tool is calculated according to the length of the cutter mark, which specifically includes: S31: According to the Pythagorean theorem: a 2 +b 2 =c 2 , calculate the distance from the cutting tool to the wafer surface, where c is the radius of the cutting tool, 2a is the length of the cutter mark, and b is the distance from the cutting tool to the wafer surface.

[0053] Specifically, the Pythagorean theorem is used to calculate the vertical distance between the tool and the wafer surface. The actual cutting depth of the tool is deduced using the known tool radius and tool mark length combined with geometric relationships, thereby determining the precise position of the tool relative to the wafer surface. This ensures that the tool can cut at the appropriate depth and "cut through" the wafer to be cut, ensuring the cutting accuracy.

[0054] S32: Calculate the difference between the radius of the cutting tool and the distance from the cutting tool to the wafer surface to obtain the actual cutting depth of the cutting tool.

[0055] Specifically, the actual cutting depth of the tool is obtained by calculating the difference between the radius of the tool and the distance from the tool to the wafer surface calculated in the previous steps. This depth value is used for subsequent tool height correction to ensure that the cutting depth during the cutting process is always accurate.

[0056] In one embodiment, if Figure 6 As shown, in step S40, the height of the cutting tool is adjusted according to the height correction value, and the moving path of the cutting tool is monitored in real time during the adjustment process of the cutting tool. If the moving path deviation is detected to exceed the preset threshold, a path correction prompt is issued, which specifically includes: S41: According to the height correction value, the tool adjustment system is controlled to adjust the cutting tool.

[0057] Specifically, according to the calculated height correction value, the adjustment instruction is transmitted to the cutting tool through the actuator control signal, and the height of the tool is adjusted so that it accurately reaches the calculated cutting depth to ensure accuracy and stability during the cutting process.

[0058] S42: During the adjustment process of the cutting tool, the moving path of the cutting tool is monitored by a motor sensor, and the deviation between the moving path of the cutting tool and a preset path is compared to obtain a moving path deviation.

[0059] Specifically, the position information of the tool is obtained through the motor position sensor, and compared with the predetermined path to calculate the path deviation value, thereby ensuring that the tool moves along the correct path and avoiding path errors. S43: When the moving path deviation exceeds a preset threshold, a path correction prompt is triggered.

[0060] Specifically, if the deviation of the tool path is detected to exceed the preset threshold, the algorithm will determine whether the deviation exceeds the allowable range. If it exceeds the range, a path correction prompt will be issued to remind the operator to make corrections or trigger the automatic correction mechanism to ensure that the tool continues to cut along the correct path.

[0061] In one embodiment, if Figure 7 As shown, the height correction method based on machine vision measurement also includes: S50: Before each cutting, the height of the cutting tool is adjusted in combination with the sensor feedback signal.

[0062] Specifically, before each cutting task begins, the sensor monitors the current height of the tool in real time, and automatically performs fine-tuning based on the feedback signal to ensure that the initial height of the tool is consistent with the predetermined target, thereby avoiding inaccurate cutting due to height errors.

[0063] S60: During the cutting process, an adaptive algorithm is used to calibrate the height of the cutting tool in real time in combination with the flatness of the cutting table where the wafer is located.

[0064] Specifically, by real-time monitoring of the flatness of the wafer cutting table, an adaptive algorithm is used to adjust the height of the cutting tool to compensate for possible minor unevenness of the cutting table, ensuring that the tool always remains in the optimal cutting position and avoiding cutting errors caused by uneven cutting table.

[0065] S70: When the wafer first direction cutting completion message is obtained, a cutting table rotation instruction is generated.

[0066] Specifically, after the cutting task in the first direction is completed, a cutting table rotation instruction is automatically generated according to the working process of the equipment, so that the cutting table rotates according to a predetermined angle to start cutting in the second direction, thereby achieving efficient cutting operation.

[0067] S80: When the message indicating that the cutting table rotation is completed is obtained, a second direction wafer position image is obtained.

[0068] Specifically, after the cutting table has completed its rotation and stabilized, a high-precision microscope camera is used to reacquire the second-direction image of the wafer to ensure that the wafer position is accurate and provide reliable image data support for cutting in the second direction.

[0069] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0070] In one embodiment, a height correction system based on machine vision measurement is provided, and the height correction system based on machine vision measurement corresponds one-to-one to the height correction method based on machine vision measurement in the above embodiment. Figure 8 As shown, the height correction system based on machine vision measurement includes a cutting module, an image acquisition module, an actual depth calculation module, a correction value calculation module and an adjustment module. The detailed description of each functional module is as follows: A cutting module, used for acquiring cutting parameter data, acquiring cutting direction data and cutting sequence data corresponding to each cutting direction data from the cutting parameter data, and generating a wafer cutting instruction according to the cutting sequence data; An image acquisition module is used to acquire a film mark image on the wafer and measure the length of the cutter mark from the film mark image using machine vision technology; The actual depth calculation module is used to calculate the actual cutting depth of the cutting tool according to the length of the cutter mark; A correction value calculation module is used to calculate the difference between the preset depth value and the actual cutting depth, calculate and determine the height correction value of the cutting tool, and dynamically adjust the height correction value according to the properties of the cutting tool. The properties of the cutting tool include material and working status; The adjustment module is used to adjust the height of the cutting tool according to the height correction value, and monitor the moving path of the cutting tool in real time during the adjustment process of the cutting tool. If the moving path deviation is detected to exceed a preset threshold, a path correction prompt is issued.

[0071] Optionally, the height correction system based on machine vision measurement also includes: A pre-adjustment module is used to adjust the height of the cutting tool before each cutting based on the sensor feedback signal; A real-time correction module is used to calibrate the height of the cutting tool in real time during the cutting process using an adaptive algorithm combined with the flatness of the cutting table where the wafer is located; A rotation module, used for generating a cutting table rotation instruction when a message that the wafer cutting in the first direction is completed is obtained; The second image acquisition module is used to acquire a wafer position image in a second direction.

[0072] Optionally, the cutting module includes: The direction determination submodule is used to extract the cutting area of ​​the wafer from the cutting parameter data and identify the cutting direction of the wafer; the sequence determination submodule is used to determine the starting point and the end point of each cutting direction based on the cutting direction, and determine the cutting order of each direction according to the cutting direction data to obtain the cutting order data; The cutting instruction generation submodule is used to generate corresponding wafer cutting instructions in combination with the cutting sequence data to control the cutting tool to perform the cutting task according to the predetermined path sequence.

[0073] Optionally, the image acquisition module includes: The shooting submodule is used to shoot the cutting marks on the wafer using microscope technology, and to extract the cutting edges of the cutting marks using image segmentation algorithms to generate film mark images.

[0074] A first cut mark acquisition submodule is used to generate a first camera movement instruction, and acquire the first cut mark endpoint coordinates according to the first camera movement instruction; The camera return submodule is used to generate a camera return instruction when the first cut endpoint information is obtained; The second cut mark acquisition submodule is used to generate a second camera movement instruction when receiving the end of the camera return instruction, and acquire the coordinates of the second cut mark endpoint according to the second camera movement instruction; The knife mark calculation submodule is used to calculate the knife mark length according to the first cut mark endpoint coordinates and the second cut mark endpoint coordinates.

[0075] Optionally, the actual depth calculation module includes: The formula calculation submodule is used to calculate the formula according to the Pythagorean theorem: a 2 +b 2 =c 2 , calculate the distance from the cutting tool to the wafer surface, where c is the radius of the cutting tool, 2a is the length of the cutter mark, and b is the distance from the cutting tool to the wafer surface; The difference calculation submodule is used to perform difference calculation on the radius of the cutting tool and the distance from the cutting tool to the wafer surface to obtain the actual cutting depth of the cutting tool.

[0076] Optionally, the adjustment module includes: The tool adjustment submodule is used to control the tool adjustment system to adjust the cutting tool according to the height correction value; The path checking submodule is used to monitor the moving path of the cutting tool through the motor sensor during the adjustment process of the cutting tool, and compare the deviation between the moving path of the cutting tool and the preset path to obtain the moving path deviation; The path correction submodule is used to trigger a path correction prompt when the moving path deviation exceeds a preset threshold.

[0077] For the specific definition of a height correction system based on machine vision measurement, please refer to the definition of a height correction method based on machine vision measurement above, which will not be repeated here. Each module in the above-mentioned height correction system based on machine vision measurement can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0078] Those skilled in the art will clearly understand that for the sake of convenience and brevity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the system can be divided into different functional units or modules to complete all or part of the functions described above.

[0079] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A height correction method based on machine vision measurement, characterized in that: The height correction method based on machine vision measurement comprises: Acquire cutting parameter data, acquire cutting direction data and cutting sequence data corresponding to each cutting direction data from the cutting parameter data, and generate a wafer cutting instruction according to the cutting sequence data; Acquire a film mark image on the wafer, and use machine vision technology to measure the length of the cutter mark of the cutting tool from the film mark image; Calculating the actual cutting depth of the cutting tool according to the length of the cutter mark; Calculating the difference between the preset depth value and the actual cutting depth, calculating and determining the height correction value of the cutting tool, and dynamically adjusting the height correction value according to the properties of the cutting tool, wherein the properties of the cutting tool include material and working status; The height of the cutting tool is adjusted according to the height correction value, and the moving path of the cutting tool is monitored in real time during the adjustment process of the cutting tool. If it is monitored that the moving path deviation exceeds a preset threshold, a path correction prompt is issued.

2. The height correction method based on machine vision measurement according to claim 1, characterized in that: The obtaining of cutting direction data and cutting sequence data corresponding to each cutting direction data from the cutting parameter data, and generating a wafer cutting instruction according to the cutting sequence data comprises: Extracting a cutting area of ​​the wafer from the cutting parameter data and identifying a cutting direction of the wafer; Based on the cutting direction, determine the starting point and the ending point of each cutting direction, and according to the data of the cutting direction, determine the cutting order of each direction to obtain the cutting order data; In combination with the cutting sequence data, the corresponding wafer cutting instruction is generated to control the cutting tool to perform the cutting task according to a predetermined path sequence.

3. The height correction method based on machine vision measurement according to claim 1 is characterized in that: The obtaining of the film mark image of the cutting tool on the wafer comprises: The cutting marks on the wafer are photographed using a microscope technique, and the cutting edges of the cutting marks are extracted using an image segmentation algorithm to generate the film mark image.

4. The height correction method based on machine vision measurement according to claim 1 is characterized in that: The method of measuring the length of the cutting tool's cut from the film cut image using machine vision technology comprises: Generate a first camera movement instruction, and obtain the first cut mark endpoint coordinates according to the first camera movement instruction; When the first cut endpoint information is obtained, a camera return instruction is generated; When receiving the camera return instruction, generating a second camera movement instruction, and acquiring the second cut mark endpoint coordinates according to the second camera movement instruction; The cut mark length is calculated based on the first cut mark endpoint coordinates and the second cut mark endpoint coordinates.

5. The height correction method based on machine vision measurement according to claim 1 is characterized in that: Calculating the actual cutting depth of the cutting tool according to the cutter mark length includes: According to the Pythagorean theorem: a 2 +b 2 =c 2 , calculate the distance from the cutting tool to the wafer surface, where c is the radius of the cutting tool, 2a is the length of the cutter mark, and b is the distance from the cutting tool to the wafer surface; The difference between the radius of the cutting tool and the distance from the cutting tool to the wafer surface is calculated to obtain the actual cutting depth of the cutting tool.

6. The height correction method based on machine vision measurement according to claim 1, characterized in that: The step of adjusting the height of the cutting tool according to the height correction value, and monitoring the moving path of the cutting tool in real time during the adjustment of the cutting tool, and issuing a path correction prompt if the moving path deviation is detected to exceed a preset threshold, includes: According to the height correction value, controlling the tool adjustment system to adjust the cutting tool; During the adjustment process of the cutting tool, the moving path of the cutting tool is monitored by a motor sensor, and the deviation between the moving path of the cutting tool and a preset path is compared to obtain the moving path deviation; When the moving path deviation exceeds the preset threshold, the path correction prompt is triggered.

7. The height correction method based on machine vision measurement according to claim 1 is characterized in that: The height correction method based on machine vision measurement also includes: Before each cutting, the height of the cutting tool is adjusted in combination with the sensor feedback signal; During the cutting process, an adaptive algorithm is used to calibrate the height of the cutting tool in real time in combination with the flatness of the cutting table where the wafer is located; When the wafer first direction cutting completion message is obtained, a cutting table rotation instruction is generated; When the message indicating that the cutting table rotation is completed is obtained, a second direction wafer position image is obtained.

8. A height correction system based on machine vision measurement, characterized in that: The height correction system based on machine vision measurement comprises: A cutting module, used for acquiring cutting parameter data, acquiring cutting direction data and cutting sequence data corresponding to each cutting direction data from the cutting parameter data, and generating a wafer cutting instruction according to the cutting sequence data; An image acquisition module is used to acquire a film mark image on the wafer, and to measure the length of a cutter mark of a cutting tool from the film mark image using machine vision technology; An actual depth calculation module is used to calculate the actual cutting depth of the cutting tool according to the length of the cutter mark; A correction value calculation module, used for calculating the difference between a preset depth value and the actual cutting depth, calculating and determining a height correction value of the cutting tool, and dynamically adjusting the height correction value according to the properties of the cutting tool, wherein the properties of the cutting tool include material and working status; The adjustment module is used to adjust the height of the cutting tool according to the height correction value, and monitor the moving path of the cutting tool in real time during the adjustment process of the cutting tool. If the moving path deviation is detected to exceed a preset threshold, a path correction prompt is issued.

Images