A size measurement method, device, lamination machine and storage medium supporting quick model change

Abstract

This invention discloses a dimensional measurement method, apparatus, stacker, and storage medium supporting rapid changeover. The invention includes: acquiring a target product image, product dimensional data, and product spacing distance of the product being changed; locating the center coordinates of each product detection area within the target product image based on a preset reference point, the product dimensional data, and the product spacing distance; identifying the target edge positions of each product detection area based on the center coordinates and the product dimensional data; and determining the dimensional measurement data corresponding to the measurement request based on the target edge positions and the product dimensional data when a measurement request is received. This invention achieves fully automated visual positioning and measurement during the changeover process, significantly shortening the changeover operation time, improving production efficiency, and enhancing the repeatability and rapid response capability of dimensional measurements. It can be widely applied in the field of machine vision dimensional inspection technology.

Description

Technical Field

[0001] This invention relates to the field of machine vision dimension inspection technology, and in particular to a dimension measurement method, apparatus, stacker, and storage medium that support rapid model changeover. Background Technology

[0002] During the lamination process, it is typically necessary to measure the dimensions of key features such as coating edges and cut edges of electrodes to ensure that product quality meets process requirements. As the lithium battery industry continues to demand higher production capacity and yield rates, the efficiency and accuracy of dimensional measurements directly impact the overall production line pace and product consistency.

[0003] Currently, in dimensional measurement systems for lamination processes, the common practice is that when production changes (e.g., switching from one electrode model to another specification), operators need to manually reposition the camera and rebuild or adjust the visual inspection model. Specific operations include manually dragging the BLOB (Binary Large Object) tool to select the product inspection area and manually setting or moving calipers to locate the target edge. This manual intervention method relies on operational experience, and the above steps must be repeated for each new product model.

[0004] However, the aforementioned existing technologies have obvious drawbacks: each time a model is changed, the shooting position and modeling need to be re-determined, which is cumbersome and time-consuming, seriously affecting production efficiency; at the same time, the method of manually adjusting vision tools is difficult to guarantee the consistency of size measurement and rapid response capability, especially in the production scenario of multiple varieties and small batches, where the problem of low measurement efficiency caused by frequent model changes is particularly prominent. Summary of the Invention

[0005] This invention provides a dimensional measurement method, apparatus, stacking machine, and storage medium that support rapid model changeover, solving the technical problems of low dimensional measurement efficiency and long production interruption time caused by the need for manual re-determination of photo positions and modeling during model changeover in existing stacking processes.

[0006] In a first aspect, the present invention provides a dimensional measurement method supporting rapid model changeover, comprising: Obtain the target product image, product size data, and product spacing distance of the current replacement product; Based on the preset reference point, the product size data, and the product interval distance, the center coordinates of the detection area of ​​each product are located from within the target product image; Based on the coordinates of each center location and the product size data, the target edge position of each product detection area is identified; When a measurement request is received, the corresponding size measurement data is determined based on the target edge position and the product size data.

[0007] In some embodiments, obtaining the target product image of the current replacement product includes: Acquire images of the currently replaced product; Calculate the gradient magnitude of the acquired image and perform non-maximum suppression to extract edge candidate points; The edge candidate points are first screened using a hysteresis threshold, which removes edge candidate points whose gradient magnitude is less than the first threshold and retains edge candidate points whose gradient magnitude is not less than the first threshold. A second step of filtering is performed on the retained edge candidate points to retain valid edge points with gradient magnitudes greater than a second threshold, as well as intermediate edge points with gradient magnitudes between a first threshold and a second threshold; wherein, the second threshold is greater than the first threshold. Perform neighborhood connectivity verification on each of the intermediate edge points, and determine the intermediate edge point that is connected to any of the valid edge points as a new valid edge point; All the effective edge points are subjected to dilation and hole filling processes to obtain a target product image with continuous edges.

[0008] In some embodiments, locating the center coordinates of each product detection area within the target product image based on a preset reference point, the product size data, and the product spacing distance includes: Extract the starting offset from the product size data; By superimposing the position coordinates of the preset reference point with the initial offset, the center position coordinates of the reference product detection area are obtained, which are used as the reference center coordinates; the reference product detection area is the product detection area located in the first column and first row of the target product image. Based on the reference center coordinates, combined with the product size data and the product interval distance, the center position coordinates of each non-reference product detection area are determined sequentially.

[0009] In some embodiments, the product size data includes electrode length and electrode width, the product spacing includes lateral spacing and longitudinal spacing, and the non-reference product detection area includes a same-row detection area, a same-column detection area, and a same-column longitudinal detection area. The step of determining the center position coordinates of each non-reference product detection area sequentially based on the reference center coordinates, combined with the product size data and the product spacing, includes: By superimposing the electrode width and the lateral spacing, a fixed lateral offset is determined; Based on the reference center coordinates, the horizontal fixed offset is accumulated sequentially to obtain the center position coordinates of each of the same row detection areas; the same row detection area is the product detection area in the target product image that is in the same row as the reference product detection area. By superimposing the electrode length and the longitudinal spacing, a fixed longitudinal offset is determined; Based on the reference center coordinates, the fixed vertical offset is accumulated sequentially to obtain the center position coordinates of each detection area in the same column; the detection area in the same column is the product detection area in the same column as the reference product detection area in the target product image. The center coordinates of each of the same row detection areas are used as the starting center coordinates. Based on the starting center coordinates, the fixed vertical offset is added sequentially to obtain the center coordinates of each column vertical area. The column vertical area is the product detection area in the same column as the same row detection area in the target product image.

[0010] In some embodiments, the target edge location includes the coating line location and the cutting line location, the product size data includes preset standard size data, preset image resolution, and associated vertical offset, and the step of identifying the target edge location of each product detection area based on the center position coordinates of each product detection area and the product size data includes: Based on the preset image resolution, the preset standard size data is scaled to determine the length of the search target; Using the center coordinates of each product detection area as the first starting position, search for the coating line position within the range indicated by the search target length; The first starting position and the associated longitudinal offset are superimposed to obtain the second starting position, and the cutting line position is searched within the range indicated by the search target length.

[0011] In some embodiments, when a measurement request is received, determining the size measurement data corresponding to the measurement request based on the target edge position and the product size data includes: When a measurement request is received, the system performs calculations based on the target edge position and the product size data, matching the measurement rules of the measurement request to determine the size measurement data. The dimensional measurement data includes at least the measured coating width and the measured dimensional deviation value.

[0012] In some embodiments, after performing the step of determining the size measurement data corresponding to the measurement request based on the target edge position and the product size data when a measurement request is received, the process includes: Calculate the deviation between the size measurement data and the preset standard size data in the product size data to obtain the measured deviation value; Obtain historical deviation values; If the measured deviation value is greater than a preset threshold or a global reset signal is received, the first weight correction algorithm is used to adaptively adjust the compensation adjustment coefficient based on the measured deviation value and the historical deviation value; the compensation adjustment coefficient is used to optimize the target edge positioning accuracy in size measurement; If the measured deviation value is less than the preset threshold or a steady-state operation signal is received, the second weight correction algorithm is used to adaptively adjust the compensation adjustment coefficient based on the measured deviation value and the historical deviation value.

[0013] Secondly, the present invention provides a dimensional measurement device supporting rapid model changeover, used to perform the aforementioned dimensional measurement method supporting rapid model changeover, comprising: The data acquisition module is used to acquire the target product image, product size data, and product spacing distance of the current replacement product; The identification and positioning module is used to locate the center coordinates of each product detection area within the target product image based on a preset reference point, the product size data, and the product interval distance. An edge recognition module is used to identify the target edge position of each product detection area based on the coordinates of each center position and the product size data; The size measurement module is used to determine the size measurement data corresponding to the measurement request based on the target edge position and the product size data when a measurement request is received.

[0014] Thirdly, the present invention provides a stacking machine, including a memory and a processor. The memory stores a computer program, and when the computer program is executed by the processor, the processor performs the steps of the above-mentioned dimensional measurement method that supports rapid changeover.

[0015] Fourthly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed, implements the above-described dimensional measurement method supporting rapid model changeover.

[0016] As can be seen from the above technical solutions, the present invention has the following advantages: The present invention provides a dimensional measurement method, device, stacker, and storage medium that supports rapid changeover. It acquires the target product image, product dimensional data, and product spacing distance of the product being changed over, and automatically locates the center coordinates of each product detection area from the target product image based on preset reference points, product dimensional data, and product spacing distance. Then, based on the center coordinates of each product detection area and product dimensional data, it automatically identifies the target edge position. Based on the target edge position and product dimensional data, it determines the dimensional measurement data required for measurement, realizing fully automated visual positioning and measurement in the changeover process. This significantly shortens the changeover operation time and improves production efficiency. At the same time, it eliminates the inconsistency caused by manual operation, which is conducive to improving the repeatability and rapid response capability of dimensional measurement. It is especially suitable for the process requirements of frequent changeovers in multi-variety, small-batch production scenarios. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 A flowchart illustrating a dimensional measurement method supporting rapid model changeover, provided as an embodiment of the present invention; Figure 2 This is a schematic diagram of the benchmark product testing area in an embodiment of the present invention; Figure 3 This is a schematic diagram of a dimensional measuring device that supports rapid model changeover, provided in an embodiment of the present invention. Figure 4 This is a schematic diagram of the hardware structure of a stacking machine provided in an embodiment of the present invention. Detailed Implementation

[0019] This invention provides a dimensional measurement method, apparatus, stacking machine, and storage medium that support rapid model changeover, to solve the technical problems of low dimensional measurement efficiency and long production interruption time caused by the need for manual re-determination of photo positions and modeling during model changeover in existing stacking processes.

[0020] To make the objectives, features, and advantages of this invention more apparent and understandable, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described below are only some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0021] Please see Figure 1 , Figure 1 An optional flowchart of a dimensional measurement method supporting rapid model changeover provided in an embodiment of the present invention is provided, the method including steps 101 to 104.

[0022] Step 101: Obtain the target product image, product size data, and product spacing distance of the current replacement product; The current replacement product refers to a specific specification product (such as a certain model of lithium battery electrode sheet) that is being measured on the lamination production line. Its size parameters and structural characteristics have been preset by the process and are completely matched with the target product model after the production line is replaced.

[0023] The target product image refers to the image obtained after edge extraction, noise reduction, and morphological processing of the acquired image of the current product, which has continuous edges, noise removal, and clearly identifiable product outlines (coating line boundaries, cutting line boundaries). It is the core image carrier for subsequent identification of the target edge position.

[0024] The product size data is a set of preset standard size parameters for the current replacement product. It includes preset benchmark size data such as product width, longitudinal dimension of coating area, and target value of coating width, as well as product inherent structural size data, such as electrode size parameters, tab parameters, and edge control range parameters. The product inherent structural size data is pre-calibrated by the production process and stored in the system database, serving as the core basis for subsequent detection area division and measurement tool parameter setting.

[0025] Product spacing refers to the fixed distance between two adjacent currently changing products on the lamination production line (horizontal / vertical along the production line conveying direction or the camera field of view). It is determined by the production process and is used to divide multiple currently changing products into independent inspection areas to avoid overlap of inspection areas for different products.

[0026] Step 101 pre-stores key parameters such as product size data and product spacing distance of each model into the database of the size measurement system, completely eliminating the debugging process of re-determining the shooting position and modeling when changing models, realizing rapid retrieval during model changeovers without the need for manual on-site calibration; at the same time, it directly connects to the subsequent inspection process, providing basic parameter support for subsequent steps such as dividing the product inspection area, identifying the target edge position, and determining size measurement data, thereby improving production and measurement efficiency.

[0027] Optionally, after the production line completes product changeover, parameter acquisition can be triggered in the following two ways: First, the operator inputs the model identifier of the current product being replaced through the human-machine interface, and the system automatically retrieves the corresponding size data and product spacing distance from the database based on the identifier. Secondly, the production line uses a model identification module (such as RFID identification or barcode scanning) to automatically identify the model of the product currently being produced and simultaneously retrieve the matching preset parameters, including size data and product spacing.

[0028] In some embodiments, obtaining the target product image of the current replacement product includes the following sub-steps: Acquire images of the currently replaced product; Calculate the gradient magnitude of the acquired image and perform non-maximum suppression to extract edge candidate points; The first step of screening edge candidate points is to use a hysteresis threshold to remove edge candidate points whose gradient magnitude is less than the first threshold and retain edge candidate points whose gradient magnitude is not less than the first threshold. A second filtering step is performed on the edge candidate points retained after the first step. Valid edge points with gradient magnitudes greater than the second threshold, and intermediate edge points with gradient magnitudes between the first and second thresholds are retained. The second threshold is greater than the first threshold. Perform neighborhood connectivity verification on each intermediate edge point, and determine the intermediate edge point that is connected to any valid edge point as the new valid edge point; Dilation and hole filling processes are applied to all valid edge points to obtain a target product image with continuous edges.

[0029] Image acquisition refers to the original image obtained by taking pictures of the current product being replaced on the lamination production line through image acquisition equipment such as industrial cameras. It contains the complete outline of at least one current product being replaced and information about its surrounding environment, and is the original data source of the target product image. Gradient magnitude refers to the intensity of gray value change between adjacent pixels in the acquired image. It reflects the degree of abrupt change in pixel gray value in the image and is the core basis for extracting edge candidate points.

[0030] Non-maximum suppression is an edge refinement method used to remove pixels where the gradient magnitude is not locally maximum, while preserving fine lines of the edge contour and improving the accuracy of edge extraction.

[0031] Edge candidate points refer to discrete pixels that are suspected to be edge points after gradient magnitude calculation and non-maximum suppression. They include real edge points (strong edge points and intermediate edge points) and interfering noise points, which need to be further screened to remove noise.

[0032] The hysteresis threshold refers to the combination of two thresholds (high threshold and low threshold) used for hierarchical screening of edge candidate points. It is the core parameter that distinguishes effective edge points, intermediate edge points and interference noise points. It is pre-stored in the visual inspection system and can be flexibly adjusted according to the specifications of the product being changed.

[0033] Valid edge points refer to discrete pixels that are confirmed to belong to the true edge of the electrode after hysteresis threshold screening and neighborhood connectivity verification. They include two types: one is strong edge points that are directly identified as being above the high threshold; the other is intermediate edge points that are between the high and low thresholds and are connected to the strong edge points. These are the core pixel points that constitute the edge of the electrode.

[0034] Intermediate edge points refer to edge candidate points whose gradient magnitudes are between the high and low thresholds of the hysteresis threshold. Whether they belong to real edges needs to be determined by neighborhood connectivity verification. They cannot be directly identified as valid edge points, nor are they directly removed as interference noise points.

[0035] Neighborhood connectivity verification refers to a verification method used to determine whether an intermediate edge point is a real edge. This embodiment adopts the eight-neighbor connectivity rule, that is, it determines whether there is an eight-neighbor connectivity relationship (top, bottom, left, right, and diagonal) between the intermediate edge point and any valid edge point (strong edge point). If there is a connection, it is identified as a real edge; otherwise, it is discarded as noise.

[0036] New valid edge points refer to pixels that are connected to valid edge points and selected from the intermediate edge points after passing the neighborhood connectivity check. These pixels are then added to the set of valid edge points to improve the integrity of the edge of the electrode.

[0037] Dilation processing refers to the morphological processing of the set of effective edge points. By pre-setting structural elements, the effective edge points are expanded to their neighborhoods, filling the tiny gaps between edge points, so that the discrete effective edge points initially form a continuous edge contour.

[0038] Hole filling refers to the subsequent morphological processing of the edge contour after expansion treatment, which fills the tiny holes and broken areas inside the edge contour, eliminates the discontinuity of the edge contour, and ensures that the edge contour is complete and smooth.

[0039] The core of acquiring the target product image is to extract clear and continuous product edges from the original acquired image. Since the original acquired image contains invalid information such as noise and background interference, it cannot be directly used for edge recognition and region segmentation. Therefore, gradient magnitude calculation is used to capture pixel gray-level abrupt change points (edge ​​candidate points). Then, non-maximum suppression is used to refine the edges, hysteresis threshold is used to filter true edges, and dilation and hole filling are used to repair the edges. Finally, a target product image with continuous edges and no noise is obtained, which provides a high-quality image foundation for subsequent region segmentation and edge localization. At the same time, this image processing flow is a preset fixed flow of the system, and no parameters need to be adjusted when changing models, which is suitable for rapid model change requirements.

[0040] As an example, the specific process for obtaining the target product image in step 101 is as follows: Step A1, perform image acquisition: use an industrial camera to take pictures of the three A-type electrodes on the production line, and obtain the acquired images containing the three electrodes to be tested. The resolution of the acquired images is 1280×720 pixels, which includes the complete outline of the electrodes, slight background noise and blurred edge areas. Step A2: Perform gradient magnitude calculation and non-maximum suppression on the acquired image obtained in step A1: Use the Sobel operator to calculate the gradient magnitude of the acquired image, capture the gray-level abrupt change points at the edge of the electrode, and obtain a set of edge candidate points; then use a 3×3 window for non-maximum suppression processing to remove pixels with local non-maximum values ​​from the edge candidate points, refine the edge contour, and remove edge coarsening interference. Step A3: Perform hysteresis thresholding on the image processed in step A2: Use the preset first threshold 40 and second threshold 80 to filter edge candidate points, retain multiple valid edge points with gray values ​​greater than the second threshold 80, and retain multiple intermediate edge points with gray values ​​between 40 and 80, while removing noise points with gray values ​​less than the first threshold 40, and initially obtain the edge ring of the electrode. Step A4: Based on the eight-neighborhood connectivity rule, perform neighborhood connectivity verification on each intermediate edge point. Identify the intermediate edge points connected to the valid edge points as new valid edge points. Perform dilation and hole filling on the image processed in Step A3: Use a 3×3 rectangular structuring element to dilate all valid edge points and connect the tiny broken parts in the edges; then use a hole filling algorithm to fill the tiny holes (holes smaller than 5 pixels) in the edge contour. Finally, obtain a target product image with continuous edges, no noise, and clearly identifiable electrode coating line boundaries and cutting line boundaries.

[0041] Step 102: Based on the preset reference point, product size data, and product interval distance, locate the center coordinates of each product detection area within the target product image; Preset reference points refer to fixed reference coordinate points that have been calibrated and stored in advance (such as the center point of the camera's field of view, fixed positioning points on the production line, etc.). These points, together with the dimensional data and product interval distance obtained in step 101, serve as the reference for dividing the product inspection area and do not require recalibration when changing models.

[0042] The product inspection area refers to the precise inspection range defined within the target product image based on preset reference points, size data, and product spacing. It has clear horizontal / vertical boundary coordinates. Its core function is to provide an independent and interference-free operating space for the size measurement of a single currently changing product, directly related to the accuracy of size measurement. Each currently changing product corresponds to an independent product inspection area, used to isolate interference from adjacent products, providing a precise, stable, and controllable dedicated inspection space for edge recognition, caliper placement, and size measurement.

[0043] The product detection area refers to the precise detection range defined in the target product image by a BLOB generation script. It has clear horizontal / vertical boundary coordinates and its core function is to provide an independent and interference-free operating space for the size measurement of a single current product, directly related to the accuracy of the size measurement. Each current product corresponds to an independent product detection area, which is used to isolate interference from adjacent products and provide a precise, stable and controllable dedicated detection space for edge recognition, caliper placement and size measurement.

[0044] A BLOB generation script is a pre-written automated program or set of rules. Its function is to automatically calculate the position parameters (such as the Y coordinate of the center point) of the BLOB region (i.e., the region of interest) of each current product in the acquired image based on the input geometric parameters (preset reference point, size data of the current replacement product, product interval distance), and delineate the corresponding BLOB region on the target edge image.

[0045] Specifically, in step 101, the target product image, product size data, and product spacing distance of the current replacement product have been obtained. Combined with the pre-stored preset reference points, there is no need to redetermine the shooting position or establish a detection model. The BLOB generation script performs BLOB pixel analysis on the target product image and automatically identifies the BLOB area corresponding to each current replacement product, i.e., the product detection area mentioned above. Since the BLOB area spans the entire camera field of view horizontally, the BLOB generation script does not need to locate the X coordinate value of the center point of the BLOB area. It only locates the Y coordinate value of the center point of each BLOB area based on the preset reference points and product size data.

[0046] In some embodiments, step 102 includes the following sub-steps: Extract the starting offset from the product size data; By superimposing the position coordinates of the preset reference point and the initial offset, the center position coordinates of the reference product detection area are obtained, which are used as the reference center coordinates; the reference product detection area is the product detection area located in the first column and first row of the target product image. Based on the reference center coordinates, combined with product size data and product spacing, the center position coordinates of each non-reference product inspection area are determined sequentially.

[0047] The initial offset refers to the offset parameter used to determine the center position of the inspection area of ​​the reference product. It is extracted from the product size data and includes the longitudinal offset (the Y-direction offset of the center of the first area relative to the preset reference point) and the lateral offset (the X-direction offset of the center of the first area relative to the preset reference point) to achieve accurate positioning of the first area.

[0048] The reference center coordinates are the center position coordinates of the reference product testing area. They are obtained by superimposing the preset reference point coordinates and the initial offset, and serve as the basic reference for locating the center coordinates of all subsequent product testing areas.

[0049] Optionally, using a preset benchmark point as a unified reference origin, the center position (benchmark center coordinates) of the benchmark product inspection area is accurately calibrated through the initial offset in the product size data, so that the first area is accurately matched with the actual position of the electrode. Then, based on the electrode's own size (electrode width, electrode length) and the distance between adjacent areas (lateral distance, longitudinal distance), the fixed lateral and longitudinal offsets are calculated. By iteratively accumulating the fixed offsets, the center coordinates of all lateral and longitudinal product inspection areas are quickly generated, realizing the synchronous positioning of multiple rows and columns of electrode sheets.

[0050] In some embodiments, product size data includes electrode length and electrode width, product spacing includes lateral spacing and longitudinal spacing, and non-reference product detection areas include same-row detection areas, same-column detection areas, and same-column longitudinal detection areas. The above determination of the center position coordinates of each non-reference product detection area, based on the reference center coordinates and combined with product size data and product spacing, includes the following sub-steps: The width of the electrode and the lateral spacing are superimposed to determine the fixed lateral offset; Based on the reference center coordinates, the fixed horizontal offset is accumulated sequentially to obtain the center position coordinates of each detection area in the same row; the detection area in the same row is the product detection area in the target product image that is in the same row as the reference product detection area. The longitudinal fixed offset is determined by superimposing the electrode length and longitudinal spacing; Based on the reference center coordinates, the fixed vertical offset is accumulated sequentially to obtain the center position coordinates of each detection area in the same column; the detection area in the same column is the product detection area in the same column as the reference product detection area in the target product image. The center coordinates of each detection area in the same row are used as the starting center coordinates. Based on the starting center coordinates, the fixed vertical offset is added sequentially to obtain the center coordinates of each vertical area in the same column. The vertical area in the same column is the product detection area in the same column as the detection area in the same row within the target product image.

[0051] In some embodiments, for example, lithium battery composite laminated electrodes are used as the current replacement product, and the specific parameters are set as follows: Product dimensions: Electrode width W=150mm, Electrode length L=120mm; Product spacing: Horizontal spacing D1=50mm, vertical spacing D2=40mm; Initial offset: Extracted from product size data, vertical offset ΔY = 30.0 mm, horizontal offset ΔX = 25.0 mm; The arrangement of the electrodes to be tested is required to be 3 columns (horizontal) × 2 rows (vertical), for a total of 6 electrodes to be tested, corresponding to 6 product testing areas; Based on the specific parameters mentioned above, the detailed process of step 102 (locating the center coordinates of the product testing area) is explained in detail, including: Step B1: Retrieve the pre-stored product size data and extract the starting offset from it. The starting offset includes the longitudinal offset (ΔY=30.0mm) and the lateral offset (ΔX=25.0mm) of the center position of the reference product inspection area relative to the preset reference point. Step B2 involves superimposing the position coordinates of the preset reference point with the initial offset to calculate the center position coordinates (reference center coordinates) of the reference product inspection area, as follows: The coordinates of the reference center X1 = the preset reference point X0 + the lateral offset ΔX = 100mm + 25.0mm = 125.0mm; The coordinates of the reference center Y1 = preset reference point Y0 + longitudinal offset ΔY = 200mm + 30.0mm = 230.0mm; The center coordinates of the benchmark product testing area are (125.0mm, 230.0mm). Step B3: Overlay the electrode width from the product size data with the lateral spacing from the product interval distance to determine the fixed lateral offset ΔX, as follows: ΔXfixed = electrode width W + lateral spacing D1 = 150mm + 50mm = 200mm; By superimposing the electrode length from the product size data and the longitudinal spacing from the product interval distance, the fixed longitudinal offset ΔY is determined as follows: ΔYfixed = electrode length L + longitudinal spacing D2 = 120mm + 40mm = 160mm; Step B4: Starting from the reference center coordinates (125.0mm, 230.0mm), the lateral fixed offset ΔXfixed = 200mm is added sequentially to obtain the center position coordinates of the two parallel detection areas one by one, as follows: First parallel inspection area: (125.0mm + 200mm, 230.0mm) = (325.0mm, 230.0mm); The second parallel inspection area: (325.0mm + 200mm, 230.0mm) = (525.0mm, 230.0mm); The three parallel detection areas are arranged side by side in the horizontal direction, with a center-to-center distance of 200mm between adjacent areas, which matches the electrode width and horizontal spacing, and there is no overlap or interference. Step B5: Starting from the reference center coordinates (125.0mm, 230.0mm), the longitudinal fixed offset ΔYfixed = 160mm is added sequentially to obtain the center position coordinates of one detection area in the same column (corresponding to the electrode in the second row of the first column), as follows: Center coordinates of the same detection area: (125.0mm, 230.0mm + 160mm) = (125.0mm, 390.0mm); This area is arranged in parallel along the longitudinal direction relative to the test area of ​​the reference product, with a center-to-center spacing of 160mm, which matches the length of the electrode and the longitudinal spacing. Step B6: Using the center coordinates of each row's detection area as the starting center coordinates, the fixed longitudinal offset ΔYfixed = 160mm is added sequentially to obtain the center coordinates of each longitudinal area in the same column (corresponding to the electrode plates in the 2nd and 3rd columns, 2nd row), as follows: The corresponding longitudinal area in the same row of the first detection area is: (325.0mm, 230.0mm + 160mm) = (325.0mm, 390.0mm). The corresponding longitudinal area in the same row of the second detection area is: (525.0mm, 230.0mm + 160mm) = (525.0mm, 390.0mm).

[0052] At this point, the center coordinates of the six product testing areas (3 columns x 2 rows) have been determined. All areas are arranged in an orderly manner along the horizontal and vertical axes, and the center coordinates accurately match the electrode size and spacing requirements, providing a stable coordinate reference for subsequent product testing area delineation and edge positioning.

[0053] Step 103: Based on the center position coordinates of each product inspection area and the product size data, identify the target edge position of each product inspection area; The target edge position refers to the coordinate position of the edge of the current replacement product in the target product image, including at least the coating line position and the cutting line position, which is the core basis for subsequent dimensional measurement data calculation.

[0054] In some embodiments, the product size data includes preset standard size data, preset image resolution, and associated vertical offset. Step 103 includes the following sub-steps: Based on the preset image resolution, the preset standard size data is scaled to determine the length of the search target; Using the center coordinates of each product's testing area as the first starting position, search for the coating line position within the range indicated by the target length. The first starting position and the associated longitudinal offset are superimposed to obtain the second starting position, and the cutting line position is searched within the range indicated by the length of the search target.

[0055] The preset standard size data is included in the size data obtained in step 101. It refers to the process reference size of the current product (such as the target value of electrode coating width, the standard value of cutting line spacing, etc.), which provides a theoretical basis for scale conversion and setting the target length for search.

[0056] The preset image resolution refers to the pixel density (unit: pixels / mm) of the image captured by the industrial camera. It is calibrated and stored in advance by the system to realize the mutual conversion between physical size (mm) and image pixel size (unit: pixels), so that the length of the search target is accurately matched with the actual product size.

[0057] The associated longitudinal offset refers to the product parameters that can be selected in the product size data. It is used to determine the second starting position of the cutting line search. Optional parameters include electrode width, tab shoulder width, etc., which can be flexibly selected according to the electrode structure and detection accuracy requirements to adapt to the process characteristics of different electrodes. Among them, the selection of the tab shoulder width parameter must meet the structural adaptation requirements of the electrode tab and the coating area to ensure the accuracy of the cutting line positioning.

[0058] Scale conversion refers to the process of converting preset standard size data (physical size) into image pixel size based on preset image resolution. The core purpose is to adapt the process reference size to the image search range parameters that can be recognized by measuring tools (calipers).

[0059] The search target length refers to the effective range (unit: pixels) of the target edge searched by the measuring tool (caliper) along the preset direction. It is determined by the preset standard size data after scale conversion and is used to constrain the edge search range to avoid missing edges or introducing irrelevant noise.

[0060] The preset direction refers to the pre-set edge search direction. Based on the edge distribution characteristics of the current product (the coating line boundary and the cutting line boundary are horizontally extended), the preset direction is horizontal (along the product width direction) to ensure accurate capture of the target edge.

[0061] Measuring tools refer to devices or tools used to identify the edge position of a target and assist in dimensional measurement. In this embodiment, a caliper tool is specifically used, and the edge search and positioning operations are driven by a script generated by the caliper.

[0062] The caliper generation script refers to a preset fixed script used to drive the caliper tool (measuring tool) to perform edge search and positioning operations. It can call the size data obtained in step 101, the Y coordinate of the center point of the BLOB area located in step 102, and the relevant parameters of the product inspection area, without needing to modify the script logic when changing models.

[0063] The first starting position refers to the center coordinates of each product inspection area, which serves as the starting reference point for coating line search. This ensures that the coating line search range accurately matches the product inspection area, providing a benchmark for the directional identification of coating lines.

[0064] The second starting position is obtained by superimposing the first starting position and the associated longitudinal offset, and serves as the starting reference point for the cutting line search, realizing the associated positioning of the cutting line and the coating line, and ensuring that the distance between the two is consistent with the preset standard size.

[0065] The coating line is the boundary between the coated and uncoated areas on the electrode. It serves as the starting reference for electrode size measurement. Its positioning accuracy directly affects the accuracy of subsequent size inspection. It needs to be accurately searched to ensure no offset, meeting the size accuracy requirements of lithium battery electrode die-cutting and stacking (the tolerance usually needs to be controlled within ±0.1mm).

[0066] The cutting line position is the boundary line position formed by the electrode cutting process. It is the termination reference for electrode size measurement. It is parallel to the coating line and the spacing is fixed. Its positioning needs to rely on the associated longitudinal offset to achieve precise correlation with the coating line, so as to ensure the overall size of the electrode is compliant.

[0067] Optionally, firstly, all edge recognition operations in step 103 are completed by driving the caliper tool through a preset caliper generation script. The caliper generation script does not require manual intervention and can automatically call various parameters obtained in the previous steps, making the operation automated and standardized, adapting to the needs of rapid model changeover, and avoiding the cumbersome process of manually debugging measurement tools in the existing technology.

[0068] Secondly, the Y-coordinate of the center point of the BLOB area provides a key basis for the positioning of the starting point of the caliper tool: In step 102, the BLOB generation script has already located the Y-coordinate value of the center point of each BLOB area (since the BLOB area runs horizontally through the entire camera field of view, there is no need to locate the X-coordinate value). The caliper generation script directly calls this Y-coordinate value and determines it as the Y-coordinate of the starting point of the caliper tool search, so that the starting point of the caliper tool search falls accurately in the longitudinal center position of the product inspection area, avoiding edge omissions or positioning deviations caused by the starting point offset.

[0069] Secondly, determining the search target length is crucial for constraining the search range of the caliper tool: the caliper generation script calls the product size data (including preset standard size data and preset image resolution) obtained in step 101, and converts the preset standard size data (physical size) into pixel size through scale conversion, thereby determining the search target length of the caliper tool; this search target length is precisely matched with the actual edge distribution range of the product, which not only enables the caliper tool to completely search the coating line position and the cutting line position, but also avoids the introduction of background noise due to an excessively large search range, thus improving the efficiency and accuracy of edge recognition.

[0070] Then, the caliper generation script drives the caliper tool, taking the Y-coordinate of the center point of the product inspection area as the starting Y-coordinate and the X-coordinate of the center position of the product inspection area as the starting X-coordinate, to determine the first starting position. It searches within the target length indication range along the preset horizontal direction (including horizontal to the right and / or horizontal to the left), and accurately finds the position of the coating line by capturing the edge gray-scale change features.

[0071] Finally, by superimposing the associated longitudinal offset (optional electrode width or electrode shoulder width) with the first starting position, the second starting position for the cutting line search is obtained, realizing the associated positioning of the cutting line and the coating line, ensuring that the relative position of the two meets the preset standard; with the second starting position as the reference, the cutting line is searched along the same preset direction within the same search target length range to complete the precise positioning of the two key edges.

[0072] In some embodiments, based on the center coordinates of the six product inspection areas determined by the above positioning, and combined with the preset standard size data, preset image resolution, and associated longitudinal offset in the product size data, the identification and positioning of the coating line and cutting line within each product inspection area are completed sequentially. The following example uses the baseline product inspection area (center coordinates (125.0 mm, 230.0 mm)) as a reference. Figure 2 , Figure 2 The above-mentioned benchmark product testing field is illustrated in detail, and the process of step 103 is explained in detail, including: Step C1: The caliper generation script automatically retrieves the preset standard size data (in this embodiment, the target value of coating width is 110mm, which is suitable for the search range requirements of coating lines and cutting lines) and the preset image resolution (1 pixel / mm) from the product size data. It then performs a scale conversion on the preset standard size data, converting the physical size into image pixel length, and determines the search target length as follows: Search target length (pixels) = preset standard size data (mm) × preset image resolution (pixels / mm) = 110mm × 1 pixel / mm = 110 pixels; Step C2, starting from the first starting position ( Figure 2 Using the green dots in the image as a reference, the location of the coating line is searched and identified, specifically including: The center position coordinates (125.0mm, 230.0mm) of the benchmark product testing area are taken as the first starting position. The corresponding image pixel coordinates are (125 pixels, 230 pixels) (because the preset image resolution is 1 pixel / mm, the physical value is consistent with the pixel value). The caliper measuring tool starts from the first starting position (125 pixels, 230 pixels) and moves along a preset direction (positive X-axis) within the range indicated by the target length (110 pixels) (i.e., the horizontal line segment from 125 pixels to 125+110=235 pixels in the X-axis direction and 230 pixels in the Y-axis direction). The Canny edge detection algorithm is used to capture edge grayscale abrupt changes to search for the coating line position. Figure 2 (The red line part in the middle) After searching and identifying, the core coordinates of the coating line were captured as (175 pixels, 230 pixels) (corresponding to physical coordinates (175.0mm, 230.0mm)). This position is the boundary line between the coating area and the non-coating area of ​​the electrode, which accurately matches the actual structure of the electrode and is located within the testing field of the benchmark product without background interference. Step C3, determine the second starting position ( Figure 2 The red dots in the image (specifically including:) Call the associated longitudinal offset in the product size data: In this embodiment, the tab shoulder width S=20mm is selected as the associated longitudinal offset (it can be switched to the tab width 150mm according to actual needs. After switching, only recalculation is required, and no other search logic needs to be adjusted). The first starting position (125 pixels, 230 pixels) is superimposed with the associated vertical offset (20 mm, corresponding to 20 pixels). Since the associated vertical offset is a horizontal offset parameter, only the X-axis coordinate is superimposed, while the Y-axis coordinate remains unchanged, to obtain the second starting position, as follows: Second starting position X coordinate = First starting position X coordinate + Associated vertical offset (pixel value) = 125 pixels + 20 pixels = 145 pixels; The second starting position Y coordinate = the first starting position Y coordinate = 230 pixels; That is, the second starting position is (145 pixels, 230 pixels) (corresponding to physical coordinates (145.0mm, 230.0mm)). This position serves as the starting reference for the cutting line search, so that the relative position of the cutting line and the coating line meets the preset requirements of the tab shoulder width. Step C4, using the second starting position as a reference, searches for and identifies the cutting line position, specifically including: The caliper measuring tool starts from the second starting position (145 pixels, 230 pixels) and searches along the same preset direction (positive X-axis direction) as the coating line. Within the range indicated by the same search target length (110 pixels) (i.e., the horizontal line segment from 145 pixels to 145+110=255 pixels in the X-axis direction and 230 pixels in the Y-axis direction), the same Canny edge detection algorithm is used to capture edge grayscale abrupt changes to search for the cut-off line position. Figure 2 (The blue line part in the middle). After searching and identifying, the core coordinates of the cutting line were captured as (285 pixels, 230 pixels) (corresponding to physical coordinates (285.0mm, 230.0mm)). This position is the boundary line formed by the electrode cutting process, which is parallel to the coating line (175 pixels, 230 pixels) and the distance between them is 110 pixels (corresponding to 110mm). This is consistent with the preset coating width target value, accurately matching the electrode size requirements and meeting the dimensional accuracy control requirements of the lithium battery electrode die-cutting process.

[0073] Repeat sub-steps C1 to C4 to sequentially identify and locate the coating lines and cutting lines in the remaining five product inspection areas. The search logic and parameter settings for all areas must be completely consistent to ensure the consistency and accuracy of batch inspection. This completes the center positioning and identification and location of the coating lines and cutting lines for all product inspection areas, providing accurate reference boundaries for subsequent electrode size measurement and defect detection.

[0074] Step 104: When a measurement request is received, determine the corresponding size measurement data based on the target edge position and product size data. Specifically, when a measurement request is received, the measurement rules of the measurement request are matched with the target edge position and product size data to perform calculations and determine the size measurement data. The dimensional measurement data includes at least the measured coating width and the measured dimensional deviation.

[0075] Measurement rules are a core component of measurement requirements, comprising two core parts: first, the calculation rules for measured dimensions (such as spacing calculation and scale conversion rules); and second, the calculation rules for measured dimension deviations (such as deviation calculation formulas and deviation judgment criteria).

[0076] The calculation and processing script retrieves preset measurement requirements and, in conjunction with the coordinate and dimensional data of the target edge position, automatically completes the entire process of numerical calculation, scale conversion, and deviation analysis according to measurement rules, without the need for manual intervention, thus improving calculation accuracy and efficiency.

[0077] The dimensional measurement data refers to the final measurement results output after the calculation and processing in step 104. It includes at least the measured coating width, the actual distance between the coating line and the cutting line, and the measured dimensional deviation value. At the same time, the target edge position coordinates can be output simultaneously to provide data support for subsequent process adjustments and quality inspection. The measured coating width refers to the actual physical dimension between the coating line boundary and the cutting line boundary obtained through the calculation process in step 104. It is calculated by combining the coordinate spacing of the target edge position with the preset image resolution and is the core indicator for measuring the precision of the electrode coating process.

[0078] The measured dimensional deviation value refers to the difference between the measured dimensions (measured coating width, actual spacing) and the preset standard dimensional data (target value of coating width, standard value of cutting line spacing). It is used to determine whether the measured dimensions meet the process requirements. The smaller the absolute value, the higher the product dimensional accuracy. The deviation calculation must follow the deviation rules of the preset measurement requirements.

[0079] Optionally, based on the target edge positions (coating line, cutting line coordinates) and product size data identified above, size measurement data is automatically determined according to preset measurement requirements. The following section uses the benchmark product inspection area in the above embodiment as an example to explain the process of step 104 in detail (the other five areas are executed according to the same logic to ensure data consistency in batch inspection), as follows: Step D1: The caliper generation script extracts the following core parameters, specifically including: Target edge location: Physical coordinates of the coating line (175.0mm, 230.0mm) and the cutting line (285.0mm, 230.0mm) in the benchmark product inspection field (pixel coordinates to physical coordinates have been converted through the preset image resolution, no additional calculation is required); Product size data: Target value of coating width (110mm) in preset standard size data, preset image resolution (1 pixel / mm, used to verify coordinate transformation accuracy); Step D2: Perform the calculations sequentially according to the preset measurement requirements and rules, as follows: Calculate the measured coating width: Since the Y coordinates of the coating line and the cutting line are the same (both are 230.0mm), the horizontal distance between them is the measured coating width. The calculation formula is: Measured coating width = Cutting line X-axis physical coordinate - Coating line X-axis physical coordinate; Substituting the data, the actual measured coating width is 285.0mm - 175.0mm = 110.02mm. Calculate the measured dimensional deviation: The calculation formula is: Measured dimensional deviation = Measured coating width - Preset coating width target value; Substituting the data, the actual dimensional deviation is calculated as follows: 110.02mm - 110mm = +0.02mm; Data verification: The measured dimensional deviation (+0.02mm) is within the preset tolerance (±0.1mm), and is therefore deemed to be dimensionally acceptable. Two decimal places are retained to improve data accuracy.

[0080] Step D3 outputs the size measurement data of the current product inspection area, specifically: measured coating width 110.02mm, measured size deviation +0.02mm, and the coordinate information of the product inspection area is associated to facilitate subsequent product traceability and sorting.

[0081] Repeat steps D1 to sub-step D3 above to determine the dimensional measurement data for the remaining 5 product inspection areas.

[0082] In some embodiments, after performing step 104, the above-described dimensional measurement method supporting rapid model changeover further includes the following steps: Calculate the deviation between the dimensional measurement data and the preset standard dimensional data in the product dimensional data to obtain the measured deviation value; Obtain historical deviation values; If the measured deviation value is greater than the preset threshold or a global reset signal is received, the first weight correction algorithm is used to adaptively adjust the compensation adjustment coefficient based on the measured deviation value and the historical deviation value; the compensation adjustment coefficient is used to optimize the target edge positioning accuracy in size measurement; If the measured deviation value is less than the preset threshold or a steady-state operation signal is received, the second weight correction algorithm is used to adaptively adjust the compensation adjustment coefficient based on the measured deviation value and the historical deviation value.

[0083] The measured deviation value refers to the difference between the measured dimensions (measured coating width, actual distance between coating line and cutting line) and the preset standard dimension data in the dimension measurement data output in step 104, which is used to optimize the compensation adjustment coefficient.

[0084] Historical deviation values ​​refer to the measured deviation values ​​obtained through the same calculation method during the dimensional measurement of previous batches of the current product model stored in the system. This forms a historical deviation database, which is used in conjunction with the current measured deviation value to achieve adaptive adjustment of the compensation adjustment coefficient.

[0085] The first weight correction algorithm is an interval gain algorithm. Its core application scenario is when the machine is changed or restarted (when a global reset signal is received). At this time, the machine is prone to large offset, resulting in large fluctuations in the deviation value. Its core logic is to divide the deviation value into different intervals and set corresponding gain coefficients (weights) for different intervals. When the deviation value is in different intervals, the corresponding weight is used to correct the compensation adjustment coefficient, so as to achieve the effect of the larger the deviation, the more accurate the adjustment range, which can quickly offset the large offset caused by machine change or restart.

[0086] The second weight correction algorithm is an exponential weighted adaptive algorithm. Its core adaptation scenario is during normal production (when a steady-state operation signal is received). At this time, the machine is prone to slow deviation of the mean due to factors such as machine wear and temperature drift. Its core logic is to assign a higher weight to the measured deviation value and assign a decreasing weight to the historical deviation value in chronological order (the weight of recent historical deviation is higher than that of the distant period). The correction amount is calculated by weighted summation, and the compensation adjustment coefficient is dynamically corrected to adapt to the slow trend of deviation change. It can offset the slow deviation caused by machine wear and temperature drift in real time.

[0087] The compensation adjustment coefficient refers to the adjustment parameter used to optimize the positioning accuracy of the target edge. It is stored in the system and can be called by the caliper generation script. It is applied to the target edge recognition process in step 103 (such as adjusting the search range and grayscale threshold of the caliper tool). By adaptively adjusting this coefficient, the edge positioning deviation can be reduced, adapting to the offset problem under different scenarios such as machine change / restart, machine wear / temperature drift, etc., improving the accuracy of size measurement, and adapting to the needs of rapid changeover without the need for manual adjustment.

[0088] Adaptive adjustment refers to the automatic adjustment of the compensation adjustment coefficient without manual intervention. It combines the measured deviation value with the historical deviation value through an adaptive weight correction algorithm, so that the compensation adjustment coefficient always adapts to the measurement accuracy requirements of the current product, realizing a closed loop of measurement → deviation → correction → optimization.

[0089] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0090] The dimension measuring device supporting rapid changeover provided in the embodiments of this application is described below. The dimension measuring device supporting rapid changeover described below and the dimension measuring method supporting rapid changeover described above can be referred to in correspondence with each other.

[0091] Reference Figure 3 , Figure 3 This is an optional structural diagram of a dimensional measurement device supporting rapid model changeover provided in an embodiment of the present invention. The device is used to implement the aforementioned dimensional measurement method supporting rapid model changeover, and may include: The data acquisition module 200 is used to acquire the target product image, product size data and product spacing distance of the current replacement product; The identification and positioning module 300 is used to locate the center coordinates of each product detection area from within the target product image based on a preset reference point, product size data, and product interval distance. The edge recognition module 400 is used to identify the target edge position of each product detection area based on the center position coordinates of each product detection area and product size data. The dimension measurement module 500 is used to determine the dimension measurement data corresponding to the measurement requirement based on the target edge position and product dimension data when a measurement request is received.

[0092] In some embodiments, the data acquisition module 200 performs the following functions: Acquire images of the currently replaced product; Calculate the gradient magnitude of the acquired image and perform non-maximum suppression to extract edge candidate points; The first step of screening edge candidate points is to use a hysteresis threshold to remove edge candidate points whose gradient magnitude is less than the first threshold and retain edge candidate points whose gradient magnitude is not less than the first threshold. A second filtering step is performed on the edge candidate points retained after the first step. Valid edge points with gradient magnitudes greater than the second threshold, and intermediate edge points with gradient magnitudes between the first and second thresholds are retained. The second threshold is greater than the first threshold. Perform neighborhood connectivity verification on each intermediate edge point, and determine the intermediate edge point that is connected to any valid edge point as the new valid edge point; Dilation and hole filling processes are applied to all valid edge points to obtain a target product image with continuous edges.

[0093] In some embodiments, the identification and positioning module 300 includes: Offset extraction unit is used to extract the starting offset from product size data; The reference determination unit is used to superimpose the position coordinates of the preset reference point and the initial offset to obtain the center position coordinates of the reference product detection area, which are used as the reference center coordinates; the reference product detection area is the product detection area located in the first column and first row of the target product image. The coordinate determination unit is used to determine the center position coordinates of each non-reference product inspection area sequentially based on the reference center coordinates, combined with product size data and product interval distance.

[0094] In some embodiments, the coordinate determination unit specifically performs the following functions: The width of the electrode and the lateral spacing are superimposed to determine the fixed lateral offset; Based on the reference center coordinates, the fixed horizontal offset is accumulated sequentially to obtain the center position coordinates of each detection area in the same row; the detection area in the same row is the product detection area in the target product image that is in the same row as the reference product detection area. The longitudinal fixed offset is determined by superimposing the electrode length and longitudinal spacing; Based on the reference center coordinates, the fixed vertical offset is accumulated sequentially to obtain the center position coordinates of each detection area in the same column; the detection area in the same column is the product detection area in the same column as the reference product detection area in the target product image. The center coordinates of each detection area in the same row are used as the starting center coordinates. Based on the starting center coordinates, the fixed vertical offset is added sequentially to obtain the center coordinates of each vertical area in the same column. The vertical area in the same column is the product detection area in the same column as the detection area in the same row within the target product image.

[0095] In some embodiments, the edge recognition module 400 specifically performs the following functions: Based on the preset image resolution, the preset standard size data is scaled to determine the length of the search target; Using the center coordinates of each product's testing area as the first starting position, search for the coating line position within the range indicated by the target length. The first starting position and the associated longitudinal offset are superimposed to obtain the second starting position, and the cutting line position is searched within the range indicated by the length of the search target.

[0096] In some embodiments, the dimension measurement module 500 specifically performs the following functions: When a measurement request is received, the system performs calculations based on the target edge location and product size data, matching the measurement rules of the measurement request to determine the size measurement data. The dimensional measurement data includes at least the measured coating width and the measured dimensional deviation.

[0097] In some embodiments, the above-described dimensional measuring device supporting rapid model changeover further includes an adaptive adjustment module, which specifically performs the following functions: Calculate the deviation between the dimensional measurement data and the preset standard dimensional data in the product dimensional data to obtain the measured deviation value; Obtain historical deviation values; If the measured deviation value is greater than the preset threshold or a global reset signal is received, the first weight correction algorithm is used to adaptively adjust the compensation adjustment coefficient based on the measured deviation value and the historical deviation value; the compensation adjustment coefficient is used to optimize the target edge positioning accuracy in size measurement; If the measured deviation value is less than the preset threshold or a steady-state operation signal is received, the second weight correction algorithm is used to adaptively adjust the compensation adjustment coefficient based on the measured deviation value and the historical deviation value.

[0098] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the above-described device and module can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0099] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or modules may be electrical, mechanical, or other forms.

[0100] The modules described as separate components may or may not be physically separate. Similarly, the components shown as modules may or may not be physical modules; they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0101] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0102] If the integrated module is implemented as a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a chip stacker to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0103] This invention also provides a stacking machine for use in the ECC measurement station of the stacking process, used for rapid changeover measurement of dimensional data such as electrode coating width. The stacking machine includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the above-mentioned dimensional measurement method that supports rapid changeover.

[0104] It is understood that the content of the above method embodiments is applicable to this device embodiment. The specific functions implemented by this device embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0105] Please see Figure 4 , Figure 4 The hardware structure of a stacking machine according to another embodiment is illustrated, including: The processor 901 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present invention. The memory 902 can be implemented as a read-only memory (ROM), static storage device, dynamic storage device, or random access memory (RAM). The memory 902 can store the operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 902 and is called and executed by the processor 901 to execute the dimension measurement method supporting rapid model changeover of the embodiments of this invention. The input / output interface 903 is used to implement information input and output; The communication interface 904 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.). Bus 905 transmits information between various components of the device (e.g., processor 901, memory 902, input / output interface 903, and communication interface 904); The processor 901, memory 902, input / output interface 903, and communication interface 904 are connected to each other within the device via bus 905.

[0106] This invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the aforementioned dimensional measurement method supporting rapid model changeover.

[0107] It is understood that the content of the above method embodiments is applicable to this storage medium embodiment. The specific functions implemented in this storage medium embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0108] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0109] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A dimensional measurement method supporting rapid model changeover, characterized in that, include: Obtain the target product image, product size data, and product spacing distance of the current replacement product; Based on the preset reference point, the product size data, and the product interval distance, the center coordinates of the detection area of ​​each product are located from within the target product image; Based on the coordinates of each center location and the product size data, the target edge position of each product detection area is identified; When a measurement request is received, the corresponding size measurement data is determined based on the target edge position and the product size data.

2. The dimensional measurement method supporting rapid model changeover according to claim 1, characterized in that, The step of obtaining the target product image of the current replacement product includes: Acquire images of the currently replaced product; Calculate the gradient magnitude of the acquired image and perform non-maximum suppression to extract edge candidate points; The edge candidate points are first screened using a hysteresis threshold, which removes edge candidate points whose gradient magnitude is less than the first threshold and retains edge candidate points whose gradient magnitude is not less than the first threshold. A second step of filtering is performed on the retained edge candidate points to retain valid edge points with gradient magnitudes greater than a second threshold, as well as intermediate edge points with gradient magnitudes between a first threshold and a second threshold; wherein, the second threshold is greater than the first threshold. Perform neighborhood connectivity verification on each of the intermediate edge points, and determine the intermediate edge point that is connected to any of the valid edge points as a new valid edge point; All the effective edge points are subjected to dilation and hole filling processes to obtain a target product image with continuous edges.

3. The dimensional measurement method supporting rapid model changeover according to claim 1, characterized in that, The step of locating the center coordinates of each product detection area within the target product image based on a preset reference point, the product size data, and the product spacing distance includes: Extract the starting offset from the product size data; By superimposing the position coordinates of the preset reference point with the initial offset, the center position coordinates of the reference product detection area are obtained, which are used as the reference center coordinates; the reference product detection area is the product detection area located in the first column and first row of the target product image. Based on the reference center coordinates, combined with the product size data and the product interval distance, the center position coordinates of each non-reference product detection area are determined sequentially.

4. The dimensional measurement method supporting rapid model changeover according to claim 3, characterized in that, The product size data includes electrode length and electrode width; the product spacing includes horizontal spacing and vertical spacing; the non-reference product inspection areas include same-row inspection areas, same-column inspection areas, and same-column vertical inspection areas; the determination of the center position coordinates of each non-reference product inspection area based on the reference center coordinates, combined with the product size data and the product spacing, includes: By superimposing the electrode width and the lateral spacing, a fixed lateral offset is determined; Based on the reference center coordinates, the horizontal fixed offset is accumulated sequentially to obtain the center position coordinates of each of the same row detection areas; the same row detection area is the product detection area in the target product image that is in the same row as the reference product detection area. By superimposing the electrode length and the longitudinal spacing, a fixed longitudinal offset is determined; Based on the reference center coordinates, the fixed vertical offset is accumulated sequentially to obtain the center position coordinates of each detection area in the same column; the detection area in the same column is the product detection area in the same column as the reference product detection area in the target product image. The center coordinates of each of the same row detection areas are used as the starting center coordinates. Based on the starting center coordinates, the fixed vertical offset is added sequentially to obtain the center coordinates of each column vertical area. The column vertical area is the product detection area in the same column as the same row detection area in the target product image.

5. The dimensional measurement method supporting rapid model changeover according to claim 1, characterized in that, The target edge position includes the coating line position and the cutting line position. The product size data includes preset standard size data, preset image resolution, and associated vertical offset. Identifying the target edge position of each product detection area based on the center coordinates of each product detection area and the product size data includes: Based on the preset image resolution, the preset standard size data is scaled to determine the length of the search target; Using the center coordinates of each product detection area as the first starting position, search for the coating line position within the range indicated by the search target length; The first starting position and the associated longitudinal offset are superimposed to obtain the second starting position, and the cutting line position is searched within the range indicated by the search target length.

6. The dimensional measurement method supporting rapid model changeover according to claim 1, characterized in that, When a measurement request is received, determining the corresponding size measurement data based on the target edge position and the product size data includes: When a measurement request is received, the system performs calculations based on the target edge position and the product size data, matching the measurement rules of the measurement request to determine the size measurement data. The dimensional measurement data includes at least the measured coating width and the measured dimensional deviation value.

7. The dimensional measurement method supporting rapid model changeover according to claim 1, characterized in that, After determining the size measurement data corresponding to the measurement request based on the target edge position and the product size data when a measurement request is received, the process includes: Calculate the deviation between the size measurement data and the preset standard size data in the product size data to obtain the measured deviation value; Obtain historical deviation values; If the measured deviation value is greater than a preset threshold or a global reset signal is received, the first weight correction algorithm is used to adaptively adjust the compensation adjustment coefficient based on the measured deviation value and the historical deviation value; the compensation adjustment coefficient is used to optimize the target edge positioning accuracy in size measurement; If the measured deviation value is less than the preset threshold or a steady-state operation signal is received, the second weight correction algorithm is used to adaptively adjust the compensation adjustment coefficient based on the measured deviation value and the historical deviation value.

8. A dimensional measuring device supporting rapid model changeover, characterized in that, A method for performing a dimensional measurement method supporting rapid model changeover as described in any one of claims 1 to 7, comprising: The data acquisition module is used to acquire the target product image, product size data, and product spacing distance of the current replacement product; The identification and positioning module is used to locate the center coordinates of each product detection area within the target product image based on a preset reference point, the product size data, and the product interval distance. An edge recognition module is used to identify the target edge position of each product detection area based on the coordinates of each center position and the product size data; The size measurement module is used to determine the size measurement data corresponding to the measurement request based on the target edge position and the product size data when a measurement request is received.

9. A stacking machine, characterized in that, The device includes a memory and a processor, wherein the memory stores a computer program that, when executed by the processor, causes the processor to perform the steps of the dimensional measurement method supporting rapid changeover as described in any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed, it implements the dimensional measurement method supporting rapid model changeover as described in any one of claims 1-7.

Images