Silicon wafer pose determination method and device, and silicon wafer pose adjustment device
By acquiring and analyzing angular region images of silicon wafers during laser scribing, and calculating rotation angles and displacements, the problem of inaccurate silicon wafer pose correction was solved, thereby improving cutting accuracy and material utilization.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- LAPLACE RENEWABLE ENERGY TECH CO LTD
- Filing Date
- 2026-02-06
- Publication Date
- 2026-06-26
AI Technical Summary
Inaccurate orientation correction of the silicon wafer during laser scribing leads to insufficient cutting precision, affecting cutting quality and material utilization.
By acquiring multiple corner region images of the silicon wafer area in the backlight image, the edge lines and corner point position information of the corner regions are determined. Combined with the position information of the image edge lines and the preset center point, the rotation angle and displacement of the silicon wafer are calculated to adjust the pose of the silicon wafer.
This technology enables precise orientation adjustment of silicon wafers, improves the cutting accuracy and material utilization of laser scribing, and meets the needs of high-efficiency photovoltaic module production.
Smart Images
Figure CN122289360A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of semiconductor or photovoltaic material processing, specifically to a method and apparatus for determining the pose of a silicon wafer and a device for adjusting the pose of a silicon wafer. Background Technology
[0002] Laser scribing machines are key equipment in the solar cell manufacturing process. They utilize high-energy-density laser beams (such as infrared or ultraviolet lasers) as "cutting tools," creating precise cuts on the surface of solar cells (or silicon wafers) through methods such as burning, vaporization, or cold peeling. This allows large, complete cells (such as the common 182 mm or 210 mm sizes) to be divided into two, three, or four smaller sub-cells. Compared to traditional mechanical scribing methods, laser scribing offers advantages such as high cutting precision (errors can be controlled within micrometers), no physical contact during the cutting process (avoiding microcracks caused by mechanical stress), and high cutting speed (suitable for large-scale mass production). It can meet the size customization requirements of photovoltaic modules with different power ratings, reduce module heat loss and improve power generation efficiency by lowering the current in sub-cells, and also allow for the rational cutting and utilization of large cells with minor defects, reducing material waste. It is an indispensable core piece of equipment in the current production of high-efficiency photovoltaic modules. Because the position of the solar cells may deviate from the predetermined position and the posture may tilt during the transportation process, the position and posture of the solar cells need to be accurately corrected before laser cutting to ensure cutting accuracy. Summary of the Invention
[0003] To address the aforementioned technical problems, this application is proposed. Embodiments of this application provide a silicon wafer pose determination method, apparatus, and silicon wafer pose adjustment device.
[0004] In a first aspect, one embodiment of this application provides a silicon wafer pose determination method, comprising: acquiring multiple corner region images of a silicon wafer region in a backlight image, wherein the backlight image is obtained by capturing images of the silicon wafer based on a backlight source and a backlight camera, and the backlight image includes the silicon wafer region; determining the corner region edge lines of each of the multiple corner region images; determining first position information of multiple first corner points of each of the multiple corner region images based on the corner region edge lines of each of the multiple corner region images, wherein the first position information is the position information of the first corner point in the coordinate system corresponding to the corner region image; determining second position information of the corresponding point of each of the multiple first corner points in the backlight image based on the multiple first position information, wherein the second position information is the position information of the first corner point in the coordinate system corresponding to the backlight image; acquiring the position information of the image edge line and a preset center point of the backlight image; and determining the rotation angle and displacement of the silicon wafer based on the second position information, the image edge line and the preset center point of the backlight image, so as to adjust the pose of the silicon wafer based on the rotation angle and displacement.
[0005] In some embodiments, determining the rotation angle and displacement of the silicon wafer based on the second position information, the image edge line of the backlight image, and the position information of the preset center point, so as to adjust the pose of the silicon wafer based on the rotation angle and displacement, includes: acquiring images of each target region in the backlight image with each corresponding point as the target center point based on the second position information; determining the position information of the second corner point in each target region image; and determining the rotation angle and displacement of the silicon wafer based on the position information of multiple second corner points, the image edge line of the backlight image, and the position information of the preset center point.
[0006] In some embodiments, determining the location information of the second corner point in each target region image includes: for each target region image, binarizing the target region image to obtain a binarized target region image; establishing a filtering template based on a preset radius; determining a heatmap based on the filtering template, the pixel values of each pixel in the binarized target region image, and a preset convolution algorithm; determining a candidate point set composed of multiple candidate pixels with pixel values of 0 in the target region image or the binarized target region image based on the pixel values of each pixel in the target region image or the binarized target region image; determining the candidate pixel with the largest heat value in the heatmap from the candidate point set, and using the location information of the candidate pixel with the largest heat value as the location information of the second corner point.
[0007] In some embodiments, before acquiring multiple corner region images of the silicon wafer region in the backlight image, the method further includes: extracting the silicon wafer region from the backlight image to obtain a silicon wafer region image; determining the position information of multiple third corner points corresponding to the silicon wafer region image; acquiring the position information of multiple preset corner points, and determining a perspective matrix based on the position information of the multiple third corner points and the position information of the multiple preset corner points; generating a corrected silicon wafer region image based on the perspective matrix and the position information of pixels in the silicon wafer region image; wherein, acquiring multiple corner region images of the silicon wafer region in the backlight image includes: acquiring multiple corner region images of the corrected silicon wafer region image.
[0008] In some embodiments, the backlight image is captured by a backlight camera during the transport of a silicon wafer along a first direction by a transport component. The backlight image also includes a transport component area, a light source area, and a background area. The light source area surrounds the silicon wafer area, the background area surrounds the light source area, the transport component area is located between the silicon wafer area and the background area, and is connected to both the silicon wafer area and the background area. The light source area includes multiple first light source sub-regions, the transport component area includes multiple transport component sub-regions, and at least one first light source sub-region is divided by at least one transport component sub-region into multiple second light source sub-regions arranged sequentially along a second direction, the second direction intersecting the first direction. Extracting the silicon wafer area from the backlight image to obtain a silicon wafer area image includes: etching the transport component area in the backlight image based on a preset etch nucleus to obtain an etched backlight image, wherein the preset etch nucleus includes multiple pixels arranged sequentially only along a third direction, the third direction intersecting the first direction, and the acute angle between the third direction and the second direction being less than 30° or the third direction being the same as the second direction; and extracting the silicon wafer area from the backlight image based on a preset contour detection algorithm to obtain a silicon wafer area image.
[0009] In some embodiments, determining the position information of multiple third corner points corresponding to a silicon wafer region image includes: extracting multiple first edge regions of the silicon wafer region image based on a preset edge detection algorithm; fitting the multiple first edge regions to silicon wafer region edge lines based on a preset straight line fitting algorithm; determining the position information of the first intersection point of every two adjacent silicon wafer region edge lines, and using the position information of the multiple first intersection points as the position information of multiple third corner points.
[0010] In some embodiments, obtaining multiple corner region images of a corrected silicon wafer region image includes: extracting multiple corner region images from the corrected silicon wafer region image based on the position information of multiple preset cropping regions, wherein the position information of the multiple cropping regions is determined based on the position information of multiple preset corner points.
[0011] In some embodiments, determining the corner region edge lines of each of the multiple corner region images includes: extracting multiple second edge regions of the corner region images based on a preset edge detection algorithm; fitting the multiple second edge regions into corner region edge lines based on a preset straight line fitting algorithm; and determining the position information of multiple first corner points of each of the multiple corner region images based on the corner region edge lines of each of the multiple corner region images, including: determining the position information of the second intersection point of every two adjacent corner region edge lines, and using the position information of the multiple second intersection points as the position information of the multiple first corner points.
[0012] In some embodiments, determining the rotation angle and displacement of a silicon wafer based on the position information of multiple second corner points, the image edge line of the backlight image, and the position information of a preset center point includes: determining the actual edge line of at least one silicon wafer region in the backlight image based on the position information of multiple second corner points; determining the rotation angle of the silicon wafer based on the actual edge line of at least one silicon wafer region and at least one image edge line; determining the position information of the center point of the silicon wafer region in the backlight image based on the position information of multiple second corner points; and determining the displacement of the silicon wafer based on the position information of the center point and the position information of the preset center point.
[0013] Secondly, one embodiment of this application provides a silicon wafer pose determination device, comprising: an acquisition module configured to acquire multiple corner region images of a silicon wafer region in a backlight image, wherein the backlight image is obtained by capturing images of the silicon wafer based on a backlight source and a backlight camera, and the backlight image includes the silicon wafer region; a first determination module configured to determine the corner region edge lines of each of the multiple corner region images; a second determination module configured to determine first position information of multiple first corner points of each of the multiple corner region images based on the corner region edge lines of each of the multiple corner region images, wherein the first position information is the position information of the first corner point in the coordinate system corresponding to the corner region image; a third determination module configured to determine second position information of corresponding points of each of the multiple first corner points in the backlight image based on the multiple first position information, wherein the second position information is the position information of the first corner point in the coordinate system corresponding to the backlight image; and a fourth determination module configured to acquire the position information of the image edge line and a preset center point of the backlight image, and determine the rotation angle and displacement of the silicon wafer based on the second position information, the image edge line and the preset center point of the backlight image, so as to adjust the pose of the silicon wafer based on the rotation angle and displacement.
[0014] Thirdly, one embodiment of this application provides a silicon wafer pose adjustment device, including: the silicon wafer pose determination device of the second aspect, configured to determine the rotation angle and displacement of the silicon wafer; and an adjustment device, configured to adjust the pose of the silicon wafer based on the rotation angle and displacement of the silicon wafer.
[0015] The silicon wafer pose determination method, apparatus, and silicon wafer pose adjustment apparatus proposed in this application first extract multiple corner region images from the silicon wafer area. Then, based on the shorter edge lines of each corner region image that are closer to the first corner point, the position information of multiple first corner points can be accurately calculated. This makes the position information of corresponding points determined based on the position information of the first corner points very close to the actual corner points of the silicon wafer. Therefore, based on the position information of multiple corresponding points, the required rotation angle and displacement of the silicon wafer can be accurately calculated, thereby precisely adjusting the silicon wafer pose and enabling the laser dicing machine to precisely cut the silicon wafer. Attached Figure Description
[0016] The above and other objects, features, and advantages of this application will become more apparent from the more detailed description of the embodiments of this application in conjunction with the accompanying drawings. The drawings are provided to further illustrate the embodiments of this application and form part of the specification. They are used together with the embodiments of this application to explain this application and do not constitute a limitation thereof. In the drawings, the same reference numerals generally represent the same components or steps.
[0017] Figure 1 The diagram shown is a schematic flowchart of a silicon wafer pose determination method provided in an exemplary embodiment of this application.
[0018] Figure 2 The image shown is a schematic diagram of a backlight image provided in an exemplary embodiment of this application.
[0019] Figure 3 The diagram shown is a structural schematic of a backlight source, a backlight camera, a transport component, and a silicon wafer provided in an exemplary embodiment of this application.
[0020] Figure 4 The image shown is a schematic diagram of a corner region image provided in an exemplary embodiment of this application.
[0021] Figure 5 The diagram shown is a flowchart illustrating a method for generating a corrected silicon wafer region image according to an exemplary embodiment of this application.
[0022] Figure 6 The image shown is a schematic diagram of a silicon wafer region provided in an exemplary embodiment of this application.
[0023] Figure 7 The image shown is a schematic diagram of a corrected silicon wafer region provided in an exemplary embodiment of this application.
[0024] Figure 8 The diagram shown is a flowchart illustrating a method for obtaining a silicon wafer region image provided in an exemplary embodiment of this application.
[0025] Figure 9 The diagram shown is a schematic diagram of the structure of a preset corrosion nucleus provided in an exemplary embodiment of this application.
[0026] Figure 10 The diagram shown is a flowchart illustrating a method for determining the location information of a third corner point according to an exemplary embodiment of this application.
[0027] Figure 11 The diagram shown is a schematic diagram of a plurality of first edge regions provided in an exemplary embodiment of this application.
[0028] Figure 12 The diagram shown is a schematic diagram of the edge lines of multiple silicon wafer regions provided in an exemplary embodiment of this application.
[0029] Figure 13 The diagram shown is a flowchart illustrating a method for determining the rotation angle and displacement of a silicon wafer according to an exemplary embodiment of this application.
[0030] Figure 14 The diagram shown is a flowchart illustrating a method for determining the rotation angle and displacement of a silicon wafer according to another exemplary embodiment of this application.
[0031] Figure 15 The diagram shown is a flowchart illustrating a specific method for determining the rotation angle and displacement of a silicon wafer according to an exemplary embodiment of this application.
[0032] Figure 16 The diagram shown is a schematic diagram of the silicon wafer pose determination device provided in an exemplary embodiment of this application.
[0033] Figure 17 The diagram shown is a schematic diagram of the silicon wafer pose adjustment device provided in an exemplary embodiment of this application.
[0034] Figure 18 The image shown is a schematic diagram of a backlight image after etching processing, provided in an exemplary embodiment of this application.
[0035] Figure 19 The diagram shown is a flowchart illustrating a method for determining the location information of a second corner point according to an exemplary embodiment of this application.
[0036] Figure label: 201, Silicon wafer area; 2011, Corner area; 2012, Edge line of silicon wafer area; 2013, Noise line; 202, Conveyor component area; 2021, Conveyor component sub-area; 203, Light source area; 2031, First light source sub-area; 2032, Second light source sub-area; 204, Background area; 301, Backlight source; 302, Backlight camera; 303, Silicon wafer; 304, Conveyor component; 1000, Silicon wafer pose determination device; 1001, Acquisition module; 1002, First determination module; 1003, Second determination module; 1004, Third determination module; 1005, Fourth determination module; 1100, Silicon wafer pose adjustment device; 1101, Adjustment device. Detailed Implementation
[0037] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.
[0038] Exemplary scenario The silicon wafer pose determination method proposed in this application can be executed by an electronic device, which can be a terminal, such as a smartphone, tablet computer, desktop computer, etc., or it can also be a server, host computer, controller, processor, etc. For example, the silicon wafer pose determination method provided in this application can be executed by a processor, which pre-stores a backlight image, or the processor can receive a backlight image captured by a backlight camera. The processor and the backlight camera can be integrated into an image processing device, or the processor and the backlight camera can be two independent devices.
[0039] Exemplary methods Figure 1 The diagram shown is a schematic flowchart of a silicon wafer pose determination method provided in an exemplary embodiment of this application. Figure 2 The image shown is a schematic diagram of a backlight image provided in an exemplary embodiment of this application. Figure 3 The diagram shown is a schematic representation of the backlight source, backlight camera, transport assembly, and silicon wafer provided in an exemplary embodiment of this application. Figure 4 The image shown is a schematic diagram of a corner region image provided in an exemplary embodiment of this application.
[0040] like Figure 1 As shown in the figure, this application provides a silicon wafer pose determination method, including the following steps 101 to 105.
[0041] Step 101: Obtain multiple corner region images of the silicon wafer area in the backlight image.
[0042] Among them, such as Figure 3 As shown, backlit image (e.g.) Figure 2 (As shown) is a backlight image of silicon wafer 303 captured by backlight source 301 and backlight camera 302, and the backlight image includes silicon wafer region 201.
[0043] For example, such as Figure 3 As shown, when capturing backlit images, the transport component 304 ( Figure 3 Only the conveyor belt is shown in the image, carrying silicon wafer 303. The transport assembly 304 and silicon wafer 303 are located between the backlight source 301 and the backlight camera 302.
[0044] For example, the backlight image of silicon wafer 303 is cropped to obtain corner region images of multiple corner regions 2011.
[0045] Step 102: Determine the edge lines of the corner regions of the multiple corner region images.
[0046] Typically, the horizontal cross-sectional shape of silicon wafer 303 is a rectangle with chamfered corners on all four sides. A rectangle with chamfered corners has eight corners in total. Corner region 2011 is the region where each of the eight corners is located. The edge lines of the corner region include multiple edge lines (or outlines) of the corner region 2011 in the corner region image. The edge lines of the corner region do not include the clipping lines in the corner region image.
[0047] For example, a corner region image such as Figure 4 As shown, Figure 4 The corner region image shown has three corner region edge lines, namely L1, L2, and L3. The corner region edge lines do not include the cropping lines of the corner region image, that is, they do not include L4 and L5.
[0048] Step 103: Based on the edge lines of the corner regions of the multiple corner region images, determine the first position information of the multiple first corner points of the multiple corner region images.
[0049] The first position information is the position information of the first corner point in the coordinate system corresponding to the corner region image.
[0050] For example, Figure 4 The angular region image shown includes two first corner points, P1 and P2.
[0051] Step 104: Based on multiple first position information, determine the second position information of the corresponding points of the multiple first corner points in the backlight image.
[0052] The second position information is the position information of the first corner point in the coordinate system corresponding to the backlight image.
[0053] The corresponding point of the first corner point in the backlight image is the corner point of the corner region of the silicon wafer region 201 in the backlight image. Since the coordinate systems of the corner region image and the backlight image are different, after calculating the first position information of multiple first corner points, it is necessary to calculate the second position information of the corresponding point of the position indicated by the first corner point in the backlight image through step 104.
[0054] Step 105: Obtain the position information of the image edge line and the preset center point of the backlight image. Based on the second position information, the position information of the image edge line and the preset center point of the backlight image, determine the rotation angle and displacement of the silicon wafer, so as to adjust the pose of the silicon wafer based on the rotation angle and displacement.
[0055] In this context, the image edge lines of the backlight image refer to the outline of the backlight image. For example, if the backlight image is a rectangular image, then the image edge lines are the four sides of the rectangular image. The preset center point position information is the standard position that the center point of the silicon wafer should reach, as preset.
[0056] In related technologies, when determining the position information of the corner points of silicon wafer region 201, multiple edge lines of silicon wafer region 201 are usually extracted directly from the backlight image. Then, the position information of the corner points of silicon wafer region 201 is determined based on the multiple edge lines of silicon wafer region 201. However, since the edge lines of silicon wafer region 201 in the captured backlight image are not straight lines, but are actually jagged and have a certain curvature, if the corner points of silicon wafer region 201 are calculated based on multiple longer edge lines of silicon wafer region 201, the position of the corner point will deviate significantly from the actual corner point of silicon wafer 303. In the above embodiment, by first extracting multiple corner region images from silicon wafer region 201, and then accurately calculating the position information of multiple first corner points based on the shorter corner region edge lines of each of the multiple corner region images that are closer to the first corner point, the position information of the corresponding point determined based on the position information of the first corner point is very close to the actual corner point of silicon wafer 303. Therefore, based on the position information of multiple corresponding points, the required rotation angle (rotation angle) and displacement of the silicon wafer 303 can be accurately calculated, thereby precisely adjusting the pose of the silicon wafer 303 so that the laser scribing machine can accurately cut the silicon wafer 303.
[0057] Figure 5 The diagram shown is a flowchart illustrating a method for generating a corrected silicon wafer region image according to an exemplary embodiment of this application. Figure 6 The image shown is a schematic diagram of a silicon wafer region provided in an exemplary embodiment of this application. Figure 7 The image shown is a schematic diagram of a corrected silicon wafer region provided in an exemplary embodiment of this application.
[0058] In some embodiments, such as Figure 5 As shown, before acquiring multiple corner region images of the silicon wafer region in the backlight image, the following steps 401 to 404 can also be performed.
[0059] Step 401: Extract the silicon wafer region from the backlight image to obtain the silicon wafer region image.
[0060] Step 402: Determine the position information of multiple third corner points corresponding to the silicon wafer region image.
[0061] Step 403: Obtain the position information of multiple preset corner points, and determine the perspective matrix based on the position information of multiple third corner points and the position information of multiple preset corner points.
[0062] Specifically, the process of step 403 is as follows.
[0063] First, a coefficient matrix can be established based on the position information of multiple third corner points and multiple preset corner points.
[0064] The coefficient matrix consists of eight rows and eight columns, and each pair of corresponding third corner points and preset corner points can construct two rows.
[0065] Second, establish a perspective matrix consisting of eight target parameters to be solved and one known parameter "1", where the parameter located in the third row and third column is 1.
[0066] For example, the perspective matrix is shown in Formula 1 below.
[0067] Formula 1: .
[0068] in, Represents the perspective matrix. , , , , , , , These represent the eight target parameters.
[0069] Third, establish the target vector.
[0070] For example, the target vector is shown in Formula 2 below.
[0071] Formula 2: .
[0072] in, Represents the target vector. and The x and y coordinates of the first preset corner point, and The x and y coordinates of the second preset corner point, and The x and y coordinates of the third preset corner point, and The x and y coordinates are the fourth preset corner point.
[0073] Fourth, based on the coefficient matrix, perspective matrix, and target vector, an equation is established to solve for the eight target parameters.
[0074] For example, the equation is shown in Formula 3 below.
[0075] Formula 3: .
[0076] in, Represents the perspective matrix. Represents the coefficient matrix. This represents the target vector.
[0077] When solving for the perspective matrix based on Formula 3, it can be done using matrix operations or the least squares method.
[0078] Step 404: Generate a corrected silicon wafer region image based on the perspective matrix and the position information of pixels in the silicon wafer region image.
[0079] For example, the corrected horizontal and vertical coordinates of pixels in a silicon wafer region image can be calculated according to the following formulas 4 and 5.
[0080] Formula 4: .
[0081] in, This represents the corrected x-coordinate of a pixel in the silicon wafer region image. This represents the x-coordinate of a pixel in the silicon wafer region image. , , , , , , , These represent the eight target parameters.
[0082] Formula 5: .
[0083] in, This represents the corrected ordinate of a pixel in the silicon wafer region image. This represents the ordinate of a pixel in a silicon wafer region image. , , , , , , , These represent the eight target parameters.
[0084] After calculating the x and y coordinates of multiple corrected pixels in the silicon wafer region image, interpolation (such as bilinear interpolation) can be performed based on the x and y coordinates of multiple corrected pixels to obtain the corrected silicon wafer region image.
[0085] After performing steps 401 to 404, when acquiring multiple corner region images of the silicon wafer region in the backlight image, multiple corner region images of the corrected silicon wafer region image can be acquired.
[0086] Because the silicon wafer area image has a certain degree of offset, it is difficult to determine the position of the corner area image without correcting the silicon wafer area image, thus making it difficult to obtain the corner area image. In the above embodiment, correcting the silicon wafer area image facilitates the subsequent acquisition of the corner area image.
[0087] In some embodiments, such as Figure 2 and Figure 3 As shown, the backlight image is captured by the backlight camera 302 during the transport of the silicon wafer 303 by the transport component 301 along the first direction (m direction in the figure). The backlight image also includes a transport component area 202, a light source area 203, and a background area 204 (such as the outer frame area of the backlight light source 301). The light source area 203 surrounds the silicon wafer area, and the background area 204 surrounds the light source area 203. The transport component area 202 is located between the silicon wafer area 201 and the background area 204 and is connected to the silicon wafer area 201 and the background area 204. The light source area 203 includes a plurality of first light source sub-areas 2031, and the transport component area 202 includes a plurality of transport component sub-areas 2021. At least one first light source sub-area 2031 is divided by at least one transport component sub-area 2021 into a plurality of second light source sub-areas 2032 arranged sequentially along the second direction (n direction in the figure). The second direction intersects the first direction.
[0088] For example, such as Figure 2 As shown, the first direction is roughly up and down, and the second direction is roughly left and right. The transport component area 202 includes four transport component sub-areas 2021, namely, two transport component sub-areas 2021 above the silicon wafer area 201 and two transport component sub-areas 2021 below the silicon wafer area 201. The light source area 203 includes four first light source sub-areas 2031, which are located on the upper, lower, left, and right sides of the silicon wafer area 201, respectively. The upper first light source sub-area 2031 is divided into three second light source sub-areas 2032 arranged in the left and right direction by the two transport component sub-areas 2021, and the lower first light source sub-area 2031 is also divided into three second light source sub-areas 2032 arranged in the left and right direction by the two transport component sub-areas 2021.
[0089] Figure 8 The diagram shown is a flowchart illustrating a method for obtaining an image of a silicon wafer region according to an exemplary embodiment of this application. Figure 9 The diagram shown is a schematic representation of a preset corrosion nucleus provided in an exemplary embodiment of this application. Figure 18 The image shown is a schematic diagram of a backlight image after etching processing, provided in an exemplary embodiment of this application.
[0090] Among them, such as Figure 8As shown, when extracting the silicon wafer region 201 from the backlight image to obtain the silicon wafer region image, the following steps 501 and 502 can be performed.
[0091] Step 501: Based on the preset corrosion nucleus, corrode the transport component area in the backlight image to obtain the backlight image after corrosion processing.
[0092] Among them, such as Figure 9 As shown, the preset erosion kernel consists of multiple pixels arranged sequentially along a third direction (as shown by direction o in the diagram). The third direction intersects the first direction, and the acute angle between the third direction and the second direction is less than 30°, or the third direction and the second direction are the same. The preset erosion kernel is a pre-defined key array or template used during the erosion operation, and can be generated using the NumPy library.
[0093] For example, the preset erosion kernel includes three pixels arranged sequentially only along a third direction (e.g., ...). Figure 9 One black square in the diagram represents one pixel.
[0094] Step 502: Based on the preset contour detection algorithm, extract the silicon wafer region from the backlight image to obtain the silicon wafer region image.
[0095] For example, the contour detection algorithm is OpenCV's findContours algorithm.
[0096] In the above embodiments, since at least one first light source sub-region 2031 is divided into a plurality of second light source sub-regions 2032 arranged sequentially along the second direction by at least one transport component sub-region 2021, the transport component sub-region 2021 between adjacent second light source sub-regions 2032 in each first light source sub-region 2031 can be precisely etched away by a preset etch kernel (unidirectional etch kernel along the third direction) with a plurality of pixels arranged sequentially along the third direction (i.e., a direction that is closer to the second direction), and the interfering background can also be precisely etched away, thereby enabling the precise extraction of the silicon wafer region image in step 502.
[0097] Figure 10 The diagram shown is a flowchart illustrating a method for determining the location information of a third corner point according to an exemplary embodiment of this application. Figure 11 The diagram shown is a schematic representation of a plurality of first edge regions provided in an exemplary embodiment of this application. Figure 12 The diagram shown is a schematic diagram of the edge lines of multiple silicon wafer regions provided in an exemplary embodiment of this application.
[0098] In some embodiments, such as Figure 10 As shown, when determining the position information of multiple third corner points corresponding to the silicon wafer area image, the following steps 601 to 603 can be performed.
[0099] Step 601: Based on the preset edge detection algorithm, extract multiple first edge regions from the silicon wafer region image.
[0100] For example, multiple first edge regions such as Figure 11 As shown.
[0101] For example, the preset edge detection algorithm is the Canny operator in OpenCV, and the first edge region obtained by this algorithm is a binary gradient map.
[0102] Step 602: Based on the preset straight line fitting algorithm, fit multiple first edge regions as edge lines of silicon wafer regions respectively.
[0103] For example, multiple silicon wafer region edge lines 2012 Figure 12 As shown (i.e.) Figure 12 (long-term) Figure 12 In addition to the silicon wafer region edge line 2012, there are also multiple noise lines 2013 (i.e., short lines connected to both ends of each silicon wafer region edge line 2012), and the multiple noise lines 2013 can be removed.
[0104] For example, multiple first edge regions can be fitted to silicon wafer region edge lines using the RANSAC algorithm in scikit-kit. The RANSAC algorithm can also remove noise lines 2013.
[0105] Step 603: Determine the location information of the first intersection point of the edge lines of each pair of adjacent silicon wafer regions, and use the location information of multiple first intersection points as the location information of multiple third corner points.
[0106] Specifically, the calculation process in step 603 is as follows.
[0107] First, determine the equation of the straight line at the edge of each pair of adjacent silicon wafer regions.
[0108] If the first edge line of two adjacent silicon wafer regions passes through point A ( , The second edge line of the silicon wafer region passes through point B. , Then, the equation of the straight line of the first silicon wafer region edge is shown in Formula 6, and the equation of the straight line of the second silicon wafer region edge is shown in Formula 7.
[0109] Formula 6: .
[0110] in, This indicates the slope of the edge line of the first silicon wafer region. Represents the vertical axis variable. This represents the x-axis variable.
[0111] Formula 7: .
[0112] in, This indicates the slope of the edge line of the second silicon wafer region. , Represents the vertical axis variable. This represents the x-axis variable.
[0113] Second, by using the equations of the straight lines of the edge lines of every two adjacent silicon wafer regions, the x and y coordinates of the first intersection point are solved.
[0114] For example, the x-coordinate and y-coordinate of the first intersection point can be solved based on the following formulas 8 and 9, respectively.
[0115] Formula 8: .
[0116] in, Represents the x-coordinate of the first intersection point. This indicates the slope of the edge line of the first silicon wafer region. This indicates the slope of the edge line of the second silicon wafer region. This represents the x-coordinate of point A. This represents the ordinate of point A. This represents the x-coordinate of point B. This represents the ordinate of point B.
[0117] After obtaining the x-coordinate of the first intersection point based on Formula 8, substitute the x-coordinate of the first intersection point into Formula 6 or Formula 7 to obtain the y-coordinate of the first intersection point. Alternatively, Formula 9 can be used to calculate the y-coordinate of the first intersection point.
[0118] Formula 9: .
[0119] in, This represents the ordinate of the first intersection point. This indicates the slope of the edge line of the first silicon wafer region. This indicates the slope of the edge line of the second silicon wafer region. This represents the x-coordinate of point A. This represents the ordinate of point A. This represents the x-coordinate of point B. This represents the ordinate of point B.
[0120] In the above embodiments, by obtaining the position information of the first intersection point of the edge lines of each two adjacent silicon wafer regions, and using the position information of multiple first intersection points as the position information of multiple third corner points, perspective transformation of the silicon wafer region image can be performed based on the position information of multiple third corner points to accurately correct the silicon wafer region image.
[0121] In some embodiments, when acquiring multiple corner region images of a rectified silicon wafer region image, multiple corner region images can be extracted from the rectified silicon wafer region image based on the position information of multiple preset cropping regions. The position information of the multiple cropping regions is determined based on the position information of multiple preset corner points.
[0122] For example, the center point of each captured region is a preset corner point.
[0123] For example, the location information of the cropped area can be determined and stored in advance based on the location information of preset corner points, or it can be determined at any time point before extracting multiple corner area images from the corrected silicon wafer area image based on the location information of multiple preset cropped areas.
[0124] In the above embodiments, by using the position information of the cropping area determined by the position information of multiple preset corner points, multiple corner area images can be extracted from the corrected silicon wafer area image, which can ensure that complete corner areas can be cropped for silicon wafers 303 of different sizes, specifications, and poses.
[0125] In some embodiments, when determining the corner region edge lines of multiple corner region images, multiple second edge regions of the corner region images can be extracted first based on a preset edge detection algorithm, and then the multiple second edge regions can be fitted into corner region edge lines based on a preset line fitting algorithm. When determining the position information of multiple first corner points of multiple corner region images based on their respective corner region edge lines, the position information of the second intersection point of every two adjacent corner region edge lines can be determined, and the position information of the multiple second intersection points can be used as the position information of the multiple first corner points.
[0126] For example, the preset edge detection algorithm is the Canny operator in OpenCV, and the second edge region obtained by this algorithm is a binarized gradient map.
[0127] For example, the preset line fitting algorithm is the RANSAC algorithm in scikit-learn.
[0128] For example, when determining the location information of the second intersection point of the edge lines of each two adjacent corner regions, the straight line equations of the edge lines of each two adjacent corner regions can be determined first. Then, the x-coordinate and y-coordinate of the second intersection point can be solved by the straight line equations of the edge lines of each two adjacent corner regions. The calculation process of this step is similar to the specific calculation process of determining the location information of the first intersection point in the aforementioned embodiment. Please refer to the specific calculation process of determining the location information of the first intersection point in the aforementioned embodiment.
[0129] In the above embodiments, this method can accurately determine the position information of multiple first corner points, so that the rotation angle and displacement of the silicon wafer 303 can be accurately calculated based on the position information of multiple first corner points in subsequent steps.
[0130] Figure 13 The diagram shown is a flowchart illustrating a method for determining the rotation angle and displacement of a silicon wafer according to an exemplary embodiment of this application.
[0131] In some embodiments, such as Figure 13 As shown, when determining the rotation angle and displacement of the silicon wafer based on the position information of multiple corresponding points, the image edge line of the backlight image, and the position information of the preset center point, so as to adjust the pose of the silicon wafer based on the rotation angle and displacement, the following steps 701 to 704 can be executed. Steps 701 and 702 are used to calculate the rotation angle of the silicon wafer, and steps 703 and 704 are used to calculate the displacement of the silicon wafer. The steps of calculating the rotation angle and calculating the displacement of the silicon wafer can be performed simultaneously or in any order.
[0132] Step 701: Based on the position information of multiple corresponding points, determine the actual edge line of at least one silicon wafer region in the backlight image.
[0133] The silicon wafer 303 has four long sides and four short sides formed by chamfers. Each short side is connected between two long sides that are perpendicular to each other in the direction of extension. In step 701, only the actual edge line of the silicon wafer area corresponding to the long side of the silicon wafer 303 can be calculated.
[0134] Specifically, there are eight corresponding points. We can first determine the two corresponding points for each long side, and then calculate the straight line equation for each long side based on the two corresponding points for each long side and the preset straight line equation algorithm.
[0135] Step 702: Determine the rotation angle of the silicon wafer based on at least one actual edge line of a silicon wafer region and at least one image edge line.
[0136] Specifically, for each image edge line, the edge angle between the actual edge line of the silicon wafer region close to each image edge line and each image edge line can be calculated, and the rotation angle of the silicon wafer 303 can be calculated based on at least one edge angle.
[0137] For example, if multiple edge angles are calculated based on multiple image edge lines and multiple actual edge lines of the silicon wafer region, the rotation angle can be obtained by averaging these multiple edge angles. If only one edge angle is calculated based on one image edge line and one actual edge line of the silicon wafer region, this single edge angle can be used as the rotation angle. It is understood that determining the rotation angle by averaging multiple edge angles has higher accuracy.
[0138] Step 703: Based on the position information of multiple corresponding points, determine the position information of the center point of the silicon wafer region in the backlight image.
[0139] For example, the location information of the center point can be obtained by averaging the location information of the eight corresponding points.
[0140] Step 704: Determine the displacement of the silicon wafer based on the position information of the center point and the position information of the preset center point.
[0141] For example, the difference between the position information of the center point and the position information of the preset center point can be calculated to obtain the displacement of the silicon wafer.
[0142] In the above embodiments, the displacement and rotation angle of the silicon wafer can be accurately calculated through steps 701 to 704.
[0143] Figure 14 The diagram shown is a flowchart illustrating a method for determining the rotation angle and displacement of a silicon wafer according to another exemplary embodiment of this application.
[0144] In some embodiments, such as Figure 14 As shown, when determining the rotation angle and displacement of the silicon wafer based on the position information of multiple corresponding points, the image edge line of the backlight image, and the position information of the preset center point, so as to adjust the pose of the silicon wafer based on the rotation angle and displacement, the following steps 801 and 802 can be performed.
[0145] Step 801: Based on the second location information, obtain images of each target region in the backlight image with each corresponding point as the target center point.
[0146] For example, the target area image can be obtained from the backlit image based on the second location information and the preset area radius.
[0147] Step 802: Determine the position information of the second corner point in the image of each target region.
[0148] Step 803: Based on the position information of multiple second corner points, the image edge line of the backlight image, and the position information of the preset center point, determine the rotation angle and displacement of the silicon wafer.
[0149] Since the position information of the first corner point is calculated based on the edge line of the corner region image, and due to the influence of the transition zone of silicon wafer region 201, the first corner point is actually not within silicon wafer region 201. For example... Figure 4 In the image, P2 represents the actual position of the first corner point, but the calculated first corner point may be at P3. Therefore, the corresponding point in the backlight image also deviates from the corner point of the silicon wafer region 201, causing the position information of the corresponding point to deviate from the actual corner point position of the silicon wafer 303. In the above embodiment, the target area image where the second corner point is located is determined by the second position information of the corresponding point, and then the position information of the second corner point is further determined from the target area image. This ensures that the position information of the second corner point is very close to the actual corner point position of the silicon wafer 303, thereby achieving pixel-level precise positioning of the silicon wafer 303.
[0150] In addition, in the process of determining the position information of the second corner point, multiple corner points (second corner point, third corner point) and intersection points (first intersection point, second intersection point) are calculated in multiple steps. This enables the accurate determination of the position information of the second corner point for silicon wafers 303 with different sizes and specifications, and also reduces the interference of camera imaging quality on the positioning results of the second corner point.
[0151] Figure 19 The diagram shown is a flowchart illustrating a method for determining the location information of a second corner point according to an exemplary embodiment of this application.
[0152] In some embodiments, such as Figure 19 As shown, when determining the location information of the second corner point in each target area image, the following steps 1201 to 1205 can be performed.
[0153] Step 1201: For each target region image, binarize the target region image to obtain the binarized target region image.
[0154] For example, the binarized target region image can be obtained according to the following formula 10.
[0155] Formula 10: , , .
[0156] in, This indicates the binarized target region image in coordinates. Pixel value at that location, Indicates the target region image in coordinates Pixel value at that location, This indicates a preset threshold, and the size of the target region image is [value missing]. .
[0157] Step 1202: Establish a filtering template based on the preset radius.
[0158] For example, a filter template can be established according to the following formula 11.
[0159] Formula 11: .
[0160] in, Indicates the filter template. This represents the coordinates of the data inside the filter template. For example, the coordinates of the first data (top left corner data) in the first row of a 3×3 filter template are (0,0). This represents the radius of the filter template.
[0161] Step 1203: Determine the heatmap based on the filter template, the pixel values of each pixel in the binarized target region image, and the preset convolution algorithm.
[0162] For example, a heat map can be determined according to the following formula 12.
[0163] Formula 12: in, Indicates the heatmap on coordinates Pixel value at that location, Indicates the radius of the filter template. Indicates the filter template in coordinates Data at the location, This represents the binarized backlight image in coordinates. The pixel value at that location.
[0164] Step 1204: Based on the pixel values of each pixel in the target region image or the binarized target region image, determine a candidate point set composed of multiple candidate pixels with a pixel value of 0 in the target region image or the binarized target region image.
[0165] For example, the candidate point set can be determined according to the following formula 13.
[0166] Formula 13: .
[0167] in, Represents the candidate point set, Indicates the target region image in coordinates The pixel value at that location, or, representing the binarized target region image at coordinates... The pixel value at that location.
[0168] Step 1205: Determine the candidate pixel with the largest heat value in the heat map from the candidate point set, and use the position information of the candidate pixel with the largest heat value as the position information of the second corner point.
[0169] For example, the position information of the second corner point can be determined according to the following formula 14.
[0170] Formula 14: in, This indicates the position information of the second corner point. Represents the candidate point set, Indicates the heatmap on coordinates The pixel value at that location.
[0171] In the above embodiments, the position information of the second corner point can be accurately determined through steps 1201 to 1205, thereby enabling pixel-level precise positioning of the silicon wafer.
[0172] Figure 15 The diagram shown is a flowchart illustrating a specific method for determining the rotation angle and displacement of a silicon wafer according to an exemplary embodiment of this application.
[0173] In some embodiments, such as Figure 15 As shown, when determining the rotation angle and displacement of the silicon wafer based on the position information of multiple second corner points, the image edge line of the backlight image, and the position information of the preset center point, the following steps 901 to 904 can be performed.
[0174] Step 901: Based on the position information of multiple second corner points, determine the actual edge line of at least one silicon wafer region in the backlight image.
[0175] The silicon wafer 303 has four long sides and four short sides formed by chamfers. Each short side is connected between two long sides that are perpendicular to each other in the direction of extension. In step 901, only the actual edge line of the silicon wafer area corresponding to the long side of the silicon wafer 303 can be calculated.
[0176] Specifically, there are eight corresponding points. We can first determine the two corresponding points for each long side, and then calculate the straight line equation for each long side based on the two corresponding points for each long side and the preset straight line equation algorithm.
[0177] Step 902: Determine the rotation angle of the silicon wafer based on at least one actual edge line of a silicon wafer region and at least one image edge line.
[0178] Specifically, for each image edge line, the edge angle between the actual edge line of the silicon wafer region close to each image edge line and each image edge line can be calculated, and the rotation angle of the silicon wafer 303 can be calculated based on at least one edge angle.
[0179] For example, if multiple edge angles are calculated based on multiple image edge lines and multiple actual edge lines of the silicon wafer region, the rotation angle can be obtained by averaging these multiple edge angles. If only one edge angle is calculated based on one image edge line and one actual edge line of the silicon wafer region, this single edge angle can be used as the rotation angle. It is understood that determining the rotation angle by averaging multiple edge angles has higher accuracy.
[0180] Step 903: Based on the position information of multiple second corner points, determine the position information of the center point of the silicon wafer region in the backlight image.
[0181] For example, the position information of the center point can be obtained by averaging the position information of the eight second corner points.
[0182] Step 904: Determine the displacement of the silicon wafer based on the position information of the center point and the position information of the preset center point.
[0183] For example, the difference between the position information of the center point and the position information of the preset center point can be calculated to obtain the displacement of the silicon wafer.
[0184] In the above embodiments, since the position information of the second corner point is very close to the position of the actual corner point of the silicon wafer 303, the rotation angle and displacement of the silicon wafer 303 calculated based on the position information of the second corner point are highly accurate. Thus, by adjusting the pose of the silicon wafer 303 based on the highly accurate rotation angle and displacement, the silicon wafer 303 can be precisely cut by the laser scribing machine.
[0185] Exemplary device Figure 16 The diagram shown is a schematic diagram of the silicon wafer pose determination device provided in an exemplary embodiment of this application.
[0186] Based on the same concept, such as Figure 16As shown in the illustration, this application embodiment also provides a silicon wafer pose determination device 1000, which includes: an acquisition module 1001, a first determination module 1002, a second determination module 1003, a third determination module 1004, and a fourth determination module 1005. The acquisition module 1001 is configured to acquire multiple corner region images of a silicon wafer region in a backlight image, wherein the backlight image is obtained by capturing images of the silicon wafer based on a backlight source and a backlight camera, and the backlight image includes the silicon wafer region. The first determination module 1002 is configured to determine the corner region edge lines of each of the multiple corner region images. The second determination module 1003 is configured to determine first position information of multiple first corner points of each of the multiple corner region images based on the corner region edge lines of each of the multiple corner region images, wherein the first position information is the position information of the first corner point in the coordinate system corresponding to the corner region image. The third determining module 1004 is configured to determine, based on multiple first position information, the second position information of the corresponding points of each of the multiple first corner points in the backlight image, wherein the second position information is the position information of the first corner point in the coordinate system corresponding to the backlight image. The fourth determining module 1004 is configured to acquire the position information of the image edge line and the preset center point of the backlight image, and based on the second position information, the position information of the image edge line and the preset center point of the backlight image, determine the rotation angle and displacement of the silicon wafer, so as to adjust the pose of the silicon wafer based on the rotation angle and displacement.
[0187] It should be understood that the description of the silicon wafer pose determination method provided in the embodiments of this application corresponds to the description of the silicon wafer pose determination device 1000. Therefore, for the parts of the silicon wafer pose determination device 1000 that are not described in detail, please refer to the description of the silicon wafer pose determination method described above.
[0188] Figure 17 The diagram shown is a schematic diagram of the silicon wafer pose adjustment device provided in an exemplary embodiment of this application.
[0189] Based on the same concept, such as Figure 17 As shown in the embodiments, this application also provides a silicon wafer pose adjustment device 1100, which includes: the silicon wafer pose determination device 1000 and the adjustment device 1101 described in the above embodiments. The silicon wafer pose determination device 1000 is configured to determine the rotation angle and displacement of the silicon wafer 303. The adjustment device 1101 is configured to adjust the pose of the silicon wafer 303 based on the rotation angle and displacement of the silicon wafer 303.
[0190] Since the silicon wafer pose adjustment device 1100 includes the silicon wafer pose determination device 1000, the silicon wafer pose adjustment device 1100 includes all the technical features and technical effects of the silicon wafer pose determination device 1000, which will not be repeated here.
[0191] The basic principles of this application have been described above with reference to specific embodiments. However, it should be noted that the advantages, benefits, and effects mentioned in this application are merely examples and not limitations, and should not be considered as essential features of each embodiment of this application. Furthermore, the specific details disclosed above are for illustrative and facilitative purposes only, and are not limitations. These details do not limit the application to the necessity of employing the aforementioned specific details for implementation.
[0192] The block diagrams of devices, apparatuses, devices, and systems involved in this application are merely illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. As those skilled in the art will recognize, these devices, apparatuses, devices, and systems can be connected, arranged, and configured in any manner. Words such as “comprising,” “including,” “having,” etc., are open-ended terms meaning “including but not limited to,” and are used interchangeably with them. The terms “or” and “and” as used herein refer to the terms “and / or,” and are used interchangeably with them unless the context clearly indicates otherwise. The term “such as” as used herein refers to the phrase “such as but not limited to,” and is used interchangeably with it.
[0193] It should also be noted that in the apparatus, equipment, and methods of this application, the components or steps can be disassembled and / or recombined. These disassemblies and / or recombinations should be considered as equivalent solutions of this application.
[0194] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use this application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein can be applied to other aspects without departing from the scope of this application. Therefore, this application is not intended to be limited to the aspects shown herein, but rather to be accorded the widest scope consistent with the principles and novel features disclosed herein.
[0195] The above description has been given for purposes of illustration and description. Furthermore, this description is not intended to limit the embodiments of this application to the forms disclosed herein. Although numerous exemplary aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, alterations, additions, and sub-combinations thereof.
Claims
1. A method for determining the pose of a silicon wafer, characterized in that, include: Acquire multiple corner region images of the silicon wafer region in a backlight image, wherein the backlight image is obtained by taking pictures of the silicon wafer based on a backlight source and a backlight camera, and the backlight image includes the silicon wafer region; Determine the corner region edge lines of each of the multiple corner region images; Based on the edge lines of the respective corner regions of the multiple corner region images, first position information of multiple first corner points of each of the multiple corner region images is determined, wherein the first position information is the position information of the first corner point in the coordinate system corresponding to the corner region image; Based on multiple first position information, second position information of each of the multiple first corner points in the backlight image is determined, wherein the second position information is the position information of the first corner point in the coordinate system corresponding to the backlight image; The position information of the image edge line and the preset center point of the backlight image is obtained. Based on the second position information and the position information of the image edge line and the preset center point of the backlight image, the rotation angle and displacement of the silicon wafer are determined, so as to adjust the pose of the silicon wafer based on the rotation angle and the displacement.
2. The silicon wafer pose determination method according to claim 1, characterized in that, The step of determining the rotation angle and displacement of the silicon wafer based on the second position information, the position information of the image edge line of the backlight image, and the position information of the preset center point, and adjusting the pose of the silicon wafer based on the rotation angle and the displacement, includes: Based on the second location information, obtain images of each target region in the backlight image with each corresponding point as the target center point; Determine the position information of the second corner point in each of the target region images; Based on the position information of multiple second corner points, the image edge line of the backlight image, and the position information of the preset center point, the rotation angle and displacement of the silicon wafer are determined.
3. The silicon wafer pose determination method according to claim 2, characterized in that, Determining the position information of the second corner point in each of the target region images includes: For each target region image, the target region image is binarized to obtain the binarized target region image; A filtering template is established based on a preset radius; A heatmap is determined based on the filtering template, the pixel values of each pixel in the binarized target region image, and a preset convolution algorithm. Based on the pixel values of each pixel in the target region image or the binarized target region image, a candidate point set is determined, consisting of multiple candidate pixels with a pixel value of 0 in the target region image or the binarized target region image. The candidate pixel with the largest heat value in the heat map is determined from the candidate point set, and the position information of the candidate pixel with the largest heat value is used as the position information of the second corner point.
4. The silicon wafer pose determination method according to claim 1 or 2, characterized in that, Before acquiring multiple corner region images of the silicon wafer region in the backlight image, the method further includes: Extract the silicon wafer region from the backlight image to obtain a silicon wafer region image; Determine the position information of multiple third corner points corresponding to the silicon wafer region image; Obtain the position information of multiple preset corner points, and determine the perspective matrix based on the position information of multiple third corner points and the position information of multiple preset corner points; Based on the perspective matrix and the position information of the pixels in the silicon wafer region image, a corrected silicon wafer region image is generated; The step of acquiring multiple corner region images of the silicon wafer region in the backlight image includes: Obtain multiple corner region images of the corrected silicon wafer region image.
5. The silicon wafer pose determination method according to claim 4, characterized in that, The backlight image is captured by the backlight camera during the transport of the silicon wafer by the transport assembly along a first direction. The backlight image also includes a transport assembly area, a light source area, and a background area. The light source area surrounds the silicon wafer area, and the background area surrounds the light source area. The transport assembly area is located between the silicon wafer area and the background area and is connected to both the silicon wafer area and the background area. The light source area includes a plurality of first light source sub-areas, and the transport assembly area includes a plurality of transport assembly sub-areas. At least one first light source sub-area is divided by at least one transport assembly sub-area into a plurality of second light source sub-areas arranged sequentially along a second direction, which intersects the first direction. The step of extracting the silicon wafer region from the backlight image to obtain a silicon wafer region image includes: The backlight image is obtained by eroding the transport component area in the backlight image based on a preset erosion kernel. The preset erosion kernel includes a plurality of pixels arranged sequentially only along a third direction. The third direction intersects the first direction, and the acute angle between the third direction and the second direction is less than 30° or the third direction is the same as the second direction. Based on a preset contour detection algorithm, the silicon wafer region in the backlight image is extracted to obtain the silicon wafer region image.
6. The silicon wafer pose determination method according to claim 4, characterized in that, The determination of the position information of multiple third corner points corresponding to the silicon wafer region image includes: Based on a preset edge detection algorithm, multiple first edge regions are extracted from the silicon wafer region image; Based on a preset straight line fitting algorithm, multiple first edge regions are respectively fitted as edge lines of silicon wafer regions; The position information of the first intersection point of each pair of adjacent silicon wafer region edge lines is determined, and the position information of multiple first intersection points is used as the position information of multiple third corner points.
7. The silicon wafer pose determination method according to claim 4, characterized in that, The acquisition of multiple corner region images of the corrected silicon wafer region image includes: Based on the position information of multiple preset cropping regions, multiple corner region images are extracted from the corrected silicon wafer region image, wherein the position information of the multiple cropping regions is determined based on the position information of multiple preset corner points.
8. The silicon wafer pose determination method according to claim 1 or 2, characterized in that, Determining the corner region edge lines of each of the plurality of corner region images includes: Based on a preset edge detection algorithm, multiple second edge regions are extracted from the corner region image; Based on a preset straight line fitting algorithm, multiple second edge regions are respectively fitted as corner region edge lines; The step of determining the position information of multiple first corner points of each of the multiple corner region images based on the edge lines of each of the multiple corner region images includes: Determine the position information of the second intersection point of the edge lines of each two adjacent corner regions, and use the position information of multiple second intersection points as the position information of multiple first corner points.
9. The silicon wafer pose determination method according to claim 2, characterized in that, The determination of the rotation angle and displacement of the silicon wafer based on the position information of multiple second corner points, the image edge line of the backlight image, and the position information of the preset center point includes: Based on the position information of multiple second corner points, determine the actual edge line of at least one silicon wafer region in the backlight image; The rotation angle of the silicon wafer is determined based on at least one actual edge line of the silicon wafer region and at least one image edge line; Based on the position information of multiple second corner points, the position information of the center point of the silicon wafer region in the backlight image is determined; The displacement of the silicon wafer is determined based on the position information of the center point and the position information of the preset center point.
10. A silicon wafer pose determination device, characterized in that, include: The acquisition module is configured to acquire multiple corner region images of the silicon wafer region in the backlight image, wherein the backlight image is obtained by taking pictures of the silicon wafer based on the backlight source and the backlight camera, and the backlight image includes the silicon wafer region; The first determining module is configured to determine the corner region edge lines of each of the plurality of corner region images; The second determining module is configured to determine first position information of a plurality of first corner points of each of the plurality of corner region images based on the edge lines of the respective corner regions of the plurality of corner region images, wherein the first position information is the position information of the first corner point in the coordinate system corresponding to the corner region image; The third determining module is configured to determine, based on the multiple first position information, the second position information of the corresponding point of each of the multiple first corner points in the backlight image, wherein the second position information is the position information of the first corner point in the coordinate system corresponding to the backlight image; The fourth determining module is configured to acquire the position information of the image edge line and the preset center point of the backlight image, and based on the second position information and the position information of the image edge line and the preset center point of the backlight image, determine the rotation angle and displacement of the silicon wafer, so as to adjust the pose of the silicon wafer based on the rotation angle and the displacement.
11. A silicon wafer pose adjustment device, characterized in that, include: The silicon wafer pose determination device of claim 10 is configured to determine the rotation angle and displacement of the silicon wafer; The adjustment device is configured to adjust the pose of the silicon wafer based on the rotation angle and the displacement of the silicon wafer.