Wafer accurate positioning method and device, electronic equipment and storage medium

Abstract

The invention belongs to the technical field of wafer positioning, and discloses a wafer accurate positioning method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining viewpoint images of different positions in a target wafer through employing a spiral collection method of a high-magnification objective lens, carrying out the image splicing of the obtained viewpoint images when each viewpoint image is obtained, and obtaining a target wafer image; the method comprises the steps of splicing viewpoint images of a wafer, carrying out mark point positioning on the spliced viewpoint images through a template matching method to obtain at least two spliced images containing preset mark points, calculating an angle offset according to position coordinates of the preset mark points to carry out angle correction, and then calculating the position offset of the wafer after angle correction to obtain an accurately positioned target wafer. The positioning precision is improved by utilizing the high-resolution characteristic of the high-magnification objective lens, the view field range of preset mark point search is expanded by utilizing the spiral multi-view image acquisition method, the objective lens does not need to be switched, the hardware of a visual imaging system is simplified, and the hardware cost of a positioning device is reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of wafer positioning, and in particular, to a wafer precise positioning method and device, an electronic device and a storage medium. BACKGROUND

[0002] In the field of semiconductor detection, for example, in the defect detection process of a MicroLED lamp bead array, due to the inevitable error of the feeding and discharging system when taking and placing the wafer, the wafer may not be in the ideal precise position after being placed. Therefore, before the subsequent defect detection, the wafer usually needs to be precisely positioned to ensure the accuracy of the detection.

[0003] The traditional wafer precise positioning method is usually based on a double objective switching mode. The low magnification objective has a larger field of view range, but its spatial resolution is relatively low, which cannot meet the demand of high-precision positioning. In order to improve the positioning accuracy, the traditional double objective switching method needs to switch to a high magnification objective to collect the positioning mark point image after completing the preliminary positioning with a low magnification. The high magnification objective has a higher spatial resolution, that is, the actual physical size corresponding to a single pixel in the image is smaller, which makes the image collected by the high magnification objective be able to provide more detailed mark point position information, thereby improving the final positioning accuracy. Although the double objective switching method can realize the precise positioning of the wafer, it has significant limitations. First, this method needs to rely on two different objectives, and the objectives need to be frequently switched during positioning. This switching mechanism makes the whole vision imaging system more complex, not only increasing the design and integration difficulty of the system, but also possibly introducing additional mechanical errors. Secondly, since two objectives and the corresponding switching mechanism need to be configured and maintained, the hardware cost of this traditional precise positioning method is usually high, which is a considerable burden for large-scale production and application.

[0004] Therefore, in order to solve the technical problems of the existing wafer precise positioning method in using the double objective switching mode, that is, the system is complex, the hardware cost is high, and the mechanical error is easy to trigger due to the frequent switching of the objectives, it is urgent to provide a wafer precise positioning method, device, electronic device and storage medium. SUMMARY

[0005] The purpose of the present application is to provide a wafer precise positioning method, device, electronic equipment and storage medium, at least two preset markers obtained by image stitching and marker positioning of the viewpoint images obtained by the spiral acquisition method are calculated to obtain the angle offset of the target wafer, so as to correct the target wafer and obtain the precisely positioned target wafer, solve the problem that the existing wafer precise positioning method uses the double objective lens switching mode, which leads to complex system, high hardware cost and easy triggering of mechanical error due to frequent switching of objective lens, only uses high magnification objective lens for spiral image acquisition, and does not need to configure and maintain two objective lenses and the corresponding switching mechanism, which significantly reduces the hardware cost and improves the positioning efficiency of the wafer.

[0006] In a first aspect, the present application provides a wafer precise positioning method for positioning a wafer, comprising the steps of: The spiral acquisition method using high magnification objective lens sequentially obtains viewpoint images at different positions in the target wafer; a plurality of preset markers for positioning are arranged in the target wafer; After each viewpoint image is obtained, the viewpoint images obtained are stitched in a nine-square form, and the marker positioning is performed on the stitched viewpoint images by a template matching method, so as to obtain at least two stitched images containing the preset markers; According to the position coordinates of the preset markers in the stitched images, the angle offset of the target wafer is calculated; Based on the angle offset, the angle of the target wafer is corrected, and the position offset of the target wafer after angle correction is calculated to obtain the precisely positioned target wafer.

[0007] The wafer precise positioning method provided by the present application can realize positioning of the wafer, at least two preset markers obtained by image stitching and marker positioning of the viewpoint images obtained by the spiral acquisition method are calculated to obtain the angle offset of the target wafer, so as to correct the angle of the target wafer and calculate the position offset of the target wafer after angle correction, and obtain the precisely positioned target wafer. The present application only uses high magnification objective lens for spiral image acquisition, on the one hand, the high resolution characteristic of the high magnification objective lens is used to improve the positioning accuracy, and on the other hand, the multi-viewpoint image acquisition method of the spiral is used to expand the field of view range of the preset marker search, without switching the objective lens. The existing wafer precise positioning method uses the double objective lens switching mode, which leads to complex system, high hardware cost and easy triggering of mechanical error due to frequent switching of objective lens, and does not need to configure and maintain two objective lenses and the corresponding switching mechanism, which significantly reduces the hardware cost and improves the positioning efficiency of the wafer.

[0008] Optionally, the spiral acquisition method using high magnification objective lens sequentially obtains viewpoint images at different positions in the target wafer, comprising: moving the high-magnification objective lens to a desired position corresponding to any preset marker point in the target wafer; According to the preset moving step, the high-magnification objective lens is controlled to move while the high-magnification objective lens is used to sequentially capture the target wafer to obtain viewpoint images at different positions in the target wafer.

[0009] The wafer precise positioning method provided in the application can position the wafer, and can ensure comprehensive and efficient collection of viewpoint images at different positions in the wafer by controlling the high-magnification objective lens to move spirally according to the preset moving step, thereby providing a complete data basis for subsequent image stitching and marker point positioning.

[0010] Optionally, after each viewpoint image is obtained, the viewpoint image obtained is stitched, and a template matching method is used to position a marker point in the stitched viewpoint image to obtain at least two stitched images containing the preset marker point, including: After each viewpoint image is obtained, the viewpoint image obtained is stitched, and a template matching method is used to position a marker point in the stitched viewpoint image to obtain at least two stitched images containing the preset marker point, including: The template matching method is used to match a marker point in each stitched viewpoint image with a preset marker point template image to detect whether any preset marker point identical to the preset marker point template image exists in each stitched viewpoint image. When it is detected that any preset marker point identical to the preset marker point template image exists in the current stitched viewpoint image, the current stitched viewpoint image is determined to be a stitched image containing the preset marker point, and the search for the current preset marker point is ended. The high-magnification objective lens is moved to a desired position corresponding to another preset marker point in the target wafer, so that the same spiral collection method, image stitching method and template matching method are used to obtain a stitched image containing another preset marker point, thereby obtaining at least two stitched images containing the preset marker point.

[0011] The wafer precise positioning method provided in the application can position the wafer, and can ensure comprehensive and efficient collection of viewpoint images at different positions in the target wafer by controlling the high-magnification objective lens to move spirally according to the preset moving step, thereby providing a complete data basis for subsequent image stitching and marker point positioning.

[0012] Optionally, after each viewpoint image is obtained, the viewpoint image obtained is stitched, and a template matching method is used to position a marker point in the stitched viewpoint image to obtain at least two stitched images containing the preset marker point, including: extracting, from the acquired view point images, adjacent images adjacent to the currently acquired view point image in eight directions according to a moving track of the high magnification objective lens each time a view point image is acquired; centering on the currently acquired view point image, stitching the currently acquired view point image and the adjacent images based on a positional relationship between the adjacent images and the currently acquired view point image to obtain a stitched view point image.

[0013] Optionally, the view point image stitched each time is matched with a preset marker point template image by a template matching method to detect whether any preset marker point in the view point image stitched each time is identical to the preset marker point template image, and the method further comprises: when the number of acquired view point images is greater than or equal to a preset maximum view point number threshold and any preset marker point in the view point image stitched each time has not been detected to be identical to the preset marker point template image, determining that a preset marker point search is timed out, terminating the search for the preset marker point, and determining that wafer positioning fails.

[0014] Optionally, the angle offset of the target wafer is calculated according to the position coordinates of the preset marker points in the stitched image, and the method comprises: calculating the position coordinates of the preset marker points according to the offset of the preset marker points from the view point center position of the corresponding stitched image in the stitched image and in combination with a preset image space resolution; inputting the position coordinates of any two preset marker points into a preset angle offset calculation formula to calculate the angle offset of the target wafer.

[0015] Optionally, the target wafer is angle-corrected based on the angle offset, and a positional offset of the target wafer after angle correction is calculated to obtain a precisely positioned target wafer, and the method comprises: angle-correcting the target wafer based on the angle offset to obtain the target wafer after angle correction; calculating a positional offset between the position coordinates of the preset marker points in the target wafer after angle correction and preset target position coordinates to provide an accurate positional deviation compensation for subsequent detection or repair, so as to obtain the precisely positioned target wafer.

[0016] In a second aspect, the application provides a wafer precise positioning device for positioning a wafer, comprising: an acquisition module configured to acquire view point images at different positions in a target wafer in sequence by using a spiral acquisition method of a high magnification objective lens; the target wafer is provided with a plurality of preset marker points for positioning. a positioning module configured to, for each acquired view image, perform image stitching on the acquired view image, and perform marker point positioning on the stitched view image by using a template matching method, to obtain at least two stitched images containing the preset marker points; a calculation module configured to calculate an angle offset of the target wafer according to the position coordinates of the preset marker points in the stitched images; a correction module configured to perform angle correction on the target wafer based on the angle offset, and calculate a position offset of the target wafer after the angle correction, to obtain the accurately positioned target wafer.

[0017] The wafer accurate positioning device calculates the angle offset of the target wafer according to the at least two preset marker points obtained by performing image stitching and marker point positioning on the view images acquired by the spiral acquisition method, performs angle correction on the target wafer, calculates the position offset of the target wafer after the angle correction, and obtains the accurately positioned target wafer. The wafer accurate positioning device solves the problems of complex system, high hardware cost and easy triggering of mechanical error caused by frequent switching of the objective lens in the existing wafer accurate positioning method using the double-objective lens switching mode. The wafer accurate positioning device only uses a high-magnification objective lens for spiral image acquisition, does not need to configure and maintain two objective lenses and the corresponding switching mechanism, significantly reduces the hardware cost, and improves the positioning efficiency of the wafer.

[0018] In a third aspect, the present application provides an electronic device, including a processor and a memory, the memory stores a computer program executable by the processor, and when the processor executes the computer program, the steps of the wafer accurate positioning method described above are run.

[0019] In a fourth aspect, the present application provides a computer readable storage medium, which stores a computer program, and when the computer program is executed by a processor, the steps of the wafer accurate positioning method described above are run. Advantageous effects

[0020] The wafer precise positioning method, device, electronic equipment and storage medium provided by the application, at least two preset marking points obtained by image stitching and marking point positioning on a viewpoint image obtained by a high magnification objective lens spiral acquisition method are calculated to obtain the angle offset of the target wafer, so as to correct the angle of the target wafer, and the position offset of the target wafer after angle correction is calculated to obtain the precisely positioned target wafer. Only high magnification objective lens is used for spiral image acquisition, on the one hand, the high resolution characteristics of the high magnification objective lens are used to improve the positioning accuracy, and on the other hand, the multi-viewpoint image acquisition method of spiral is used to expand the field of view range of the preset marking point search, without switching the objective lens. The existing wafer precise positioning method solves the problems of complex system, high hardware cost and easy triggering of mechanical error caused by frequent switching of the objective lens in the double objective lens switching mode, and does not need to configure and maintain two objective lenses and the corresponding switching mechanism, which significantly reduces the hardware cost and improves the positioning efficiency of the wafer. BRIEF DESCRIPTION OF DRAWINGS

[0021] Figure 1 The flowchart of the wafer precise positioning method provided by the embodiment of the application.

[0022] Figure 2 The structure diagram of the wafer precise positioning device provided by the embodiment of the application.

[0023] Figure 3 The structure diagram of the electronic equipment provided by the embodiment of the application.

[0024] Figure 4 The schematic diagram of the initial positioning of the preset marking point by the high magnification objective lens.

[0025] Figure 5 The schematic diagram of the spiral movement track in the spiral acquisition method.

[0026] Figure 6 The schematic diagram of the preset marking point in the spliced image.

[0027] Figure 7 The schematic diagram of the target wafer relative to the ideal position of the wafer.

[0028] Figure 8 The schematic diagram of the precisely positioned target wafer relative to the ideal position of the wafer.

[0029] Label explanation: 1, acquisition module; 2, positioning module; 3, calculation module; 4, correction module; 301, processor; 302, memory; 303, communication bus. DETAILED DESCRIPTION

[0030] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the drawings herein can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the claimed present application, but only represents selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of the present application.

[0031] It should be noted that: similar reference numerals and letters represent similar items in the following drawings, therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. Meanwhile, in the description of the present application, the terms "first", "second", etc. are only used to distinguish the description, and cannot be understood as indicating or implying relative importance.

[0032] Please refer to Figure 1 , Figure 1 is a wafer precise positioning method in some embodiments of the present application, which is used for positioning a wafer, comprising: Step S101, a spiral acquisition method of a high magnification objective lens is adopted to sequentially acquire viewpoint images of different positions in a target wafer; a plurality of preset mark points for positioning are arranged in the target wafer; Step S102, after acquiring each viewpoint image, the acquired viewpoint image is spliced in a nine-square grid splicing form, and the spliced viewpoint image is positioned by a mark point through a template matching method, to obtain at least two spliced images containing the preset mark points; Step S103, the angle offset of the target wafer is calculated according to the position coordinates of the preset mark points in the spliced images; Step S104, the target wafer is angle-corrected based on the angle offset, and the position offset of the angle-corrected target wafer is calculated, to obtain a precisely positioned target wafer.

[0033] This wafer precision positioning method calculates the angular offset of the target wafer by performing image stitching and marker point positioning on viewpoint images acquired using a high-magnification objective lens spiral acquisition method, thereby performing angle correction on the target wafer and calculating the position offset of the target wafer after angle correction to obtain a precisely positioned target wafer. By using only a high-magnification objective lens for spiral image acquisition, the high-resolution characteristics of the high-magnification objective lens are utilized to improve positioning accuracy, while the spiral multi-viewpoint image acquisition method is utilized to expand the field of view of the preset marker point search without switching the objective lens. This method solves the problems of existing wafer precision positioning methods that require frequent switching of objective lenses when using a dual-objective lens switching method, resulting in complex systems, high hardware costs, and easy triggering of mechanical errors. Furthermore, there is no need to configure and maintain two objective lenses and corresponding switching mechanisms, significantly reducing hardware costs and improving wafer positioning efficiency.

[0034] Specifically, in step S101, a spiral acquisition method using a high-magnification objective lens is used to sequentially acquire viewpoint images at different positions in the target wafer, including: Move the high-magnification objective lens to the ideal position corresponding to any preset mark point on the target wafer; A spiral acquisition method is used to control the movement of the high-magnification objective lens according to the preset moving step size, and the target wafer is photographed in sequence using the high-magnification objective lens to obtain viewpoint images at different positions in the target wafer.

[0035] In step S101, the high-magnification objective lens is accurately moved to the ideal position corresponding to any preset mark point on the target wafer (i.e., the position of the preset mark point when the target wafer is at the ideal position), ensuring that the starting point of subsequent image acquisition has a clear reference benchmark. Figure 4 As shown, Figure 4 The figure is a schematic diagram of the positioning of the preset mark point of the high-magnification objective lens, where d is the preset mark point, the circle e is the target wafer, the rectangle L in the circle e is the lamp array, f is the high-magnification objective lens, g is the imaging lens, h is the image sensor, and the arrow k is the imaging direction. The high-magnification objective lens f, the imaging lens g, and the image sensor h constitute the visual imaging system, as shown in Figure 2. Figure 4 As shown, the high-magnification objective lens f is initially positioned to an ideal position of any preset mark point d as the acquisition starting point.

[0036] By controlling the high-magnification objective lens to move spirally according to the preset moving step (the preset moving step includes a horizontal coordinate moving step and a vertical coordinate moving step, the horizontal coordinate moving step is used for moving in the horizontal coordinate direction, the vertical coordinate moving step is used for moving in the vertical coordinate direction, the horizontal coordinate moving step and the vertical coordinate moving step can be moved according to actual needs, and the horizontal coordinate moving step and the vertical coordinate moving step can be set to the same or different values according to the number of pixels in the image width and height directions multiplied by the spatial resolution, that is , , the horizontal coordinate moving step, N is the number of width pixels of the image shot by the high-magnification objective lens, and R is the spatial resolution of the image shot by the high-magnification objective lens, the vertical coordinate moving step, M is the number of height pixels of the image shot by the high-magnification objective lens) and synchronously shooting (as shown in Figure 5 , Figure 5 is a schematic diagram of a moving track of the spiral movement in the spiral acquisition method, the circles a1, …, and a25 are optical centers when the high-magnification objective lens moves each time (the number after a represents the moving order), the line segment b is a moving track of the high-magnification objective lens, and the dashed rectangle c is a view point area of an uncollected image), which can systematically and comprehensively cover each area of the target wafer and ensure that a sufficient number and quality of view point images are obtained, which are crucial for subsequent image stitching and marker positioning.

[0037] Before moving the high-magnification objective lens relative to the target wafer, the target wafer needs to be placed on the base of the three-axis motion table, a global coordinate system is established with the coordinate of the center of the target wafer in the ideal position (i.e. the position after accurate positioning) as the origin, the z-axis of the global coordinate system coincides with the rotation axis of the motion table, and the plane where the x-axis and the y-axis of the global coordinate system are located is parallel to the target wafer on the base of the three-axis motion table. When positioning the target wafer, the three-axis motion table can drive the target wafer to rotate along the z-axis, or translate along the x-axis direction and the y-axis direction to change the position coordinate of the target wafer.

[0038] The spiral acquisition method refers to that the high-magnification objective lens moves in a spiral path in the clockwise or counterclockwise direction above the target wafer according to the preset moving step, and at the same time, the target wafer is intermittently shot (i.e. shot once each time), so as to obtain view point images covering different areas of the target wafer.

[0039] The preset marker refers to a specific pattern or structure pre-set on the target wafer for assisting positioning, which can be a cross marker, a circular marker or a two-dimensional code, and the ideal position and features thereof are known.

[0040] Specifically, in step S102, each time a viewpoint image is acquired, the acquired viewpoint image is stitched in a nine-square grid stitching manner, and a marker point positioning is performed on the stitched viewpoint image by a template matching method, to obtain at least two stitched images containing the preset marker points, including: Each time a viewpoint image is acquired, the acquired viewpoint image is stitched with the adjacent acquired viewpoint images as the center, to obtain the stitched viewpoint image in a nine-square grid stitching manner; The template matching method is used to match the preset marker point template image with each stitched viewpoint image, to detect whether any preset marker point identical to the preset marker point template image exists in each stitched viewpoint image; When it is detected that any preset marker point identical to the preset marker point template image exists in the current stitched viewpoint image, the current stitched viewpoint image is determined as the stitched image containing the preset marker point, and the search for the current preset marker point is ended. The high-magnification objective lens is moved to the ideal position corresponding to another preset marker point in the target wafer, to acquire the stitched image containing another preset marker point by the same spiral acquisition method, image stitching method and template matching method, to obtain at least two stitched images containing the preset marker points.

[0041] Specifically, in step S102, each time a viewpoint image is acquired, the acquired viewpoint image is stitched with the adjacent acquired viewpoint images as the center, to obtain the stitched viewpoint image in a nine-square grid stitching manner, including: When each viewpoint image is acquired, the adjacent images adjacent to the current acquired viewpoint image in eight directions are extracted from the acquired viewpoint images according to the moving track of the high-magnification objective lens; The current acquired viewpoint image is stitched with the adjacent images based on the positional relationship between the adjacent images and the current acquired viewpoint image, to obtain the stitched viewpoint image.

[0042] In step S102, during the wafer positioning process, a single viewpoint image may not be able to completely cover the preset marker point, so it is necessary to expand the field of view by image stitching to ensure that the preset marker point can be completely captured. Each time a new viewpoint image is generated, the images having a spatial adjacency relationship with the current acquired viewpoint image in eight directions are selected from the acquired viewpoint images as the adjacent images according to the moving direction and historical track of the high-magnification objective lens.

[0043] With the current acquired viewpoint image as the center, based on the positional relationship between the adjacent image and the current acquired viewpoint image, the current acquired viewpoint image and the adjacent image are spliced to form a larger spliced image, and the spliced viewpoint image is obtained. Through this splicing method, only the current viewpoint image and the determined adjacent image are spliced locally each time, rather than all the collected images are spliced globally, so as to realize the purpose of reducing the calculation complexity and improving the real-time splicing.

[0044] In some optional embodiments, when splicing images, not only the nine-square splicing form can be used, but also other matrix arrangement methods can be used for splicing, such as QxQ or QxP (Q and P are positive integers) splicing form. The nearby acquired images are spliced with the current acquired viewpoint image as the center to obtain the spliced viewpoint.

[0045] In step S102, whenever a new viewpoint image is acquired and spliced to obtain a spliced viewpoint image, a template matching (such as existing matching calculation methods based on normalized cross-correlation correlation coefficient method, contour shape-based recognition algorithm, local feature point matching algorithm (such as SIFT, ORB), or machine learning driven target detection method) is performed immediately to match the preset marker point template image with the spliced viewpoint image each time, so as to detect the occurrence of the preset marker point in the spliced viewpoint image in real time. This real-time detection mechanism ensures that once the condition of the occurrence of the preset marker point is met during the image splicing process, the current spliced viewpoint image can be determined as the target image (i.e., the spliced image containing the preset marker point) immediately, and the search for the current preset marker point is ended, thereby avoiding unnecessary image acquisition and splicing and improving the efficiency. The existing matching calculation method is prior art, which will not be described in detail here.

[0046] Through the above method, only a high-magnification objective lens is used, images are sequentially acquired at multiple viewpoints around the ideal position of the preset marker point according to the multi-viewpoint image acquisition strategy, and the multi-viewpoint images are spliced and fused. The method of fusing “small images (viewpoint images)” into a large image (spliced image) enables the preset marker point to enter the field of view, fully utilizes the high spatial resolution characteristics of the high-magnification objective lens, acquires high-precision positioning data of the preset marker point and the wafer, and can simplify the hardware of the visual imaging system while maintaining high positioning accuracy, thereby reducing the hardware cost of the positioning device.

[0047] Specifically, in step S102, after the marker point matching between the spliced viewpoint image each time and the preset marker point template image is performed by the template matching method to detect whether any preset marker point identical to the preset marker point template image exists in the spliced viewpoint image each time, the method further includes: When the number of acquired viewpoint images is greater than or equal to the preset maximum viewpoint number threshold, and it is not detected that any preset marker point in the viewpoint image after each splicing is the same as the preset marker point template image, it is determined that the preset marker point search is timed out, the search for the preset marker point is terminated, and it is determined that the wafer positioning fails, In step S102, the number of acquired viewpoint images is counted in real time during the image acquisition process. When the number of viewpoint images reaches the preset maximum viewpoint number threshold, and the preset marker point is not detected, it is determined that the current wafer positioning process has exceeded the normal time consumption range, the marker point search process is terminated, and invalid resource consumption is avoided.

[0048] In some optional embodiments, after a wafer positioning fails once, another preset marker point can be selected for repositioning. If positioning failures occur continuously, a worker is notified in time for abnormal processing.

[0049] In step S102, after the first preset marker point is detected, the position of the high-magnification objective lens is adjusted to the ideal position of another preset marker point, and the spiral acquisition method, the image splicing method, and the template matching method are repeated to quickly lock the position of the other preset marker point (if the number of acquired viewpoint images is greater than or equal to the preset maximum viewpoint number threshold, the other preset marker point is selected for repositioning), and at least two spliced images containing the preset marker point are obtained.

[0050] Specifically, in step S103, the angle offset of the target wafer is calculated according to the position coordinates of the preset marker points in the spliced images, including: The position coordinates of the preset marker points are calculated according to the offset of the preset marker points from the viewpoint center position of the corresponding spliced image and the preset image spatial resolution. The position coordinates of the preset marker points are calculated according to the offset of the preset marker points from the viewpoint center position of the corresponding spliced image and the preset image spatial resolution.

[0051] In step S103, the pixel offset of the preset marker point relative to the viewpoint center position is determined according to the pixel number difference of the preset marker point from the viewpoint center position in the horizontal coordinate direction and the pixel number difference in the vertical coordinate direction in the spliced image. Then, the pixel offset is multiplied by the preset image spatial resolution to calculate the offset coordinates of the preset marker point relative to the viewpoint center position. The offset coordinates and the coordinates of the viewpoint center position are combined to calculate the accurate position coordinates of the preset marker point on the wafer plane. That is, the position coordinates of the preset marker point can be calculated by the following marker point coordinate formula: ; ; wherein, is the horizontal coordinate of the preset marker point; is the horizontal coordinate of the preset marker point corresponding to the center position of the view point of the spliced image; n is the pixel offset of the preset marker point relative to the center position of the view point in the horizontal coordinate direction; and R is the preset image space resolution; is the vertical coordinate of the preset marker point; is the vertical coordinate of the preset marker point corresponding to the center position of the view point of the spliced image; and m is the pixel offset of the preset marker point relative to the center position of the view point in the vertical coordinate direction.

[0052] As shown in Figure 6 , Figure 6 is a schematic view of the preset marker point in the spliced image, wherein the circle a1, the circle a4, the circle a5, the circle a6, the circle a7, the circle a8, the circle a18, the circle a19, and the circle a20 are the optical centers of the high magnification objective lens each time of movement, the solid line area p is the spliced image of the view point area that has been collected, the circle a6 is the center position of the view point of the spliced image, the dashed line rectangle c is the view point area of the image that has not been collected (in the nine-square image splicing, if the images of the adjacent view points have not been collected, all the pixel points of the image corresponding to the view point area are assigned a value of 0), d is the preset marker point, x1 is the horizontal coordinate of the offset coordinate of the preset marker point relative to the center position of the view point, and y1 is the vertical coordinate of the offset coordinate of the preset marker point relative to the center position of the view point. It can be known from Figure 6 that the horizontal coordinate x1 and the vertical coordinate y1 calculated by the pixel offset and the image space resolution, in combination with the actual coordinates of the center position of the view point of the spliced image, can be used to calculate the horizontal coordinate and the vertical coordinate of the preset marker point d.

[0053] Based on the calculated position coordinates of the preset marker point, the position coordinates of any two preset marker points are input into a preset angle offset calculation formula to calculate the angle offset of the target wafer.

[0054] wherein the preset angle offset calculation formula is specifically: ; wherein, is the angle offset; is a two-parameter arctangent function, and the angle output range thereof is ; is the horizontal coordinate of one of the preset marker points; is the vertical coordinate of one of the preset marker points; is the horizontal coordinate of the other preset marker point; is the vertical coordinate of the other preset marker point; An angle formed by a line connecting the two preset mark points (one of the preset mark points and the other preset mark point) and a positive direction axis of the horizontal coordinate axis when the target wafer is in the ideal position, that is, , is a vertical coordinate of one of the preset mark points when in the ideal position, is a horizontal coordinate of one of the preset mark points when in the ideal position, is a vertical coordinate of the other preset mark point when in the ideal position, is a horizontal coordinate of the other preset mark point when in the ideal position.

[0055] Specifically, in step S104, based on the angle offset, the target wafer is angle-corrected, and a position offset of the angle-corrected target wafer is calculated, to obtain a precisely positioned target wafer, including: Based on the angle offset, the target wafer is angle-corrected to obtain an angle-corrected target wafer; The position offset between the position coordinates of the preset mark points in the angle-corrected target wafer and the preset target position coordinates is calculated, to provide an accurate position deviation compensation for subsequent detection or repair, and to obtain a precisely positioned target wafer.

[0056] In step S104, the three-axis motion platform is rotated relative to the vision imaging system by an angle (counterclockwise rotation is positive, and clockwise rotation is negative) based on the previously calculated angle offset, to adjust the target wafer, so that the overall angle of the target wafer is aligned with the ideal state (ideal position), to obtain an angle-corrected target wafer, wherein the position coordinates of the preset mark points are angle-corrected following the rotation of the target wafer, and the position coordinates of the preset mark points after angle correction are: ; ; wherein, is a horizontal coordinate of the preset mark point after angle correction; is a horizontal coordinate of the preset mark point before angle correction; is a vertical coordinate of the preset mark point after angle correction; is a vertical coordinate of the preset mark point before angle correction.

[0057] As shown in Figure 7 , the angle offset is a horizontal coordinate of the target wafer relative to the ideal position of the wafer, and the ideal position of the wafer is a horizontal coordinate of the target wafer relative to the ideal position of the wafer. Figure 7 is a schematic diagram of the target wafer relative to the ideal position of the wafer, wherein, is an angle offset, the solid circle e is a target wafer, the dashed line e' is an ideal position of a wafer, d1 is one of the preset mark points, and d2 is the other preset mark point. From Figure 7As can be seen from the figure, the target wafer is rotated- Angle, the angle of the target wafer e can be corrected.

[0058] After completing the angle correction, the position offset of the target wafer in the X and Y directions is calculated by comparing the actual position coordinates of the preset mark point after angle correction with its ideal position coordinates, and the target wafer is accurately positioned. For example, if the actual position of the preset mark point after angle correction is (x', y'), and the ideal position is (x t ,y t ), then the position offset is (Δx,Δy) = (x t -x', y t -y'). This position offset can provide precise offset compensation for subsequent inspection or repair, ensuring the accuracy of the inspection or repair. When the subsequent inspection equipment or repair equipment performs operations, it can fine-tune its own motion trajectory or action point based on this compensation, thereby ensuring that its operation can be accurately aligned with the target area on the wafer.

[0059] like Figure 8 As shown, Figure 8 Schematic diagram of the target wafer after angle correction relative to its ideal position, where the solid circle e is the target wafer, the dotted line e' is the ideal position of the wafer, d3 is one of the preset marking points, d4 is another preset marking point, d3' is the ideal position of one of the preset marking points, and d4' is the ideal position of the other preset marking point. Figure 8 It can be seen from FIG that the offset between the actual position coordinates of the preset marking point and its ideal position can be determined to determine the offset between the target wafer e and the ideal wafer position e′.

[0060] It can be seen from the above that the wafer precise positioning method acquires the viewpoint images at different positions in the target wafer in sequence by adopting the spiral acquisition method of the high-magnification objective lens, the target wafer is provided with a plurality of preset mark points for positioning, each time a viewpoint image is acquired, the acquired viewpoint image is spliced in the form of a nine-square grid, and the mark point positioning is performed on the spliced viewpoint image by the template matching method, at least two spliced images containing the preset mark points are obtained, the angle offset of the target wafer is calculated according to the position coordinates of the preset mark points in the spliced images, the angle correction is performed on the target wafer based on the angle offset, the position offset of the target wafer after the angle correction is calculated, and the precisely positioned target wafer is obtained. Only the high-magnification objective lens is used for spiral image acquisition, on the one hand, the high resolution characteristics of the high-magnification objective lens are used to improve the positioning precision, and on the other hand, the multi-viewpoint image acquisition method of the spiral type is used to expand the field of view range of the preset mark point search, without the need to switch the objective lens. The existing wafer precise positioning method solves the problems of complex system, high hardware cost and easy triggering of mechanical errors caused by frequent switching of the objective lens when the double-objective lens switching mode is used, and without the need to configure and maintain two objective lenses and the corresponding switching mechanism, the hardware cost is significantly reduced, and the positioning efficiency of the wafer is improved.

[0061] Reference Figure 2 The application provides a wafer precise positioning device for positioning a wafer, comprising: An acquisition module 1 is configured to acquire viewpoint images at different positions in a target wafer in sequence by adopting a spiral acquisition method of a high-magnification objective lens; the target wafer is provided with a plurality of preset mark points for positioning; A positioning module 2 is configured to, each time a viewpoint image is acquired, splice the acquired viewpoint image in the form of a nine-square grid, and perform mark point positioning on the spliced viewpoint image by a template matching method, to obtain at least two spliced images containing the preset mark points; A calculation module 3 is configured to calculate the angle offset of the target wafer according to the position coordinates of the preset mark points in the spliced images; A correction module 4 is configured to perform angle correction on the target wafer based on the angle offset, calculate the position offset of the target wafer after the angle correction, and obtain the precisely positioned target wafer.

[0062] The wafer precision positioning device calculates the angular offset of the target wafer by performing image stitching and marker point positioning on viewpoint images acquired by a high-magnification objective lens spiral acquisition method, thereby performing angle correction on the target wafer and calculating the position offset of the target wafer after angle correction to obtain a precisely positioned target wafer. By using only a high-magnification objective lens for spiral image acquisition, the high-resolution characteristics of the high-magnification objective lens are utilized to improve positioning accuracy, while the spiral multi-viewpoint image acquisition method is utilized to expand the field of view of the preset marker point search without switching the objective lens. This solves the problems of existing wafer precision positioning methods that require frequent switching of objective lenses when using a dual-objective lens switching method, resulting in complex systems, high hardware costs, and easy triggering of mechanical errors. Furthermore, there is no need to configure and maintain two objective lenses and corresponding switching mechanisms, significantly reducing hardware costs and improving wafer positioning efficiency.

[0063] Specifically, when the acquisition module 1 adopts a spiral acquisition method using a high-magnification objective lens to sequentially acquire viewpoint images at different positions in the target wafer, it executes: Move the high-magnification objective lens to the ideal position corresponding to any preset mark point on the target wafer; A spiral acquisition method is used to control the movement of the high-magnification objective lens according to the preset moving step size, and the target wafer is photographed in sequence using the high-magnification objective lens to obtain viewpoint images at different positions in the target wafer.

[0064] When the acquisition module 1 is executed, the high-magnification objective lens is accurately moved to the ideal position corresponding to any preset mark point on the target wafer (that is, the position of the preset mark point when the target wafer is in the ideal position), ensuring that the starting point of subsequent image acquisition has a clear reference benchmark. Figure 4 As shown, Figure 4 Figure 1 is a schematic diagram of the positioning of the preset mark point of the high-magnification objective lens, where d is the preset mark point, e is the target wafer, f is the high-magnification objective lens, g is the imaging lens, h is the image sensor, and arrow k is the imaging direction. The high-magnification objective lens f, imaging lens g, and image sensor h constitute a visual imaging system, as shown in Figure 1. Figure 4 As shown, the high-magnification objective lens f is initially positioned to an ideal position of any preset mark point d as the acquisition starting point.

[0065] By controlling the high-magnification objective lens to move in accordance with a preset step length (the preset step length includes a horizontal coordinate movement step length and a vertical coordinate movement step length, wherein the horizontal coordinate movement step length is used for movement in the horizontal coordinate direction, and the vertical coordinate movement step length is used for movement in the vertical coordinate direction, the horizontal coordinate movement step length and the vertical coordinate movement step length can be moved according to actual needs, and the horizontal coordinate movement step length and the vertical coordinate movement step length can be set to the same or different values ​​according to the number of pixels in the image width and height directions multiplied by the spatial resolution, that is, , wherein, is the horizontal coordinate moving step, N is the number of width pixels of the high magnification objective lens captured image, and R is the spatial resolution of the high magnification objective lens captured image, is the vertical coordinate moving step, M is the number of height pixels of the high magnification objective lens captured image), and the spiral movement is performed synchronously with the photographing (as shown in Figure 5 , and Figure 5 is a schematic diagram of the moving track of the spiral movement in the spiral acquisition method, wherein the circles a1, …, a25 are the optical centers of the high magnification objective lens at each movement (the number after a represents the movement sequence), the line segment b is the moving track of the high magnification objective lens, and the dashed rectangle c is the view point area of the unacquired image), which can systematically and comprehensively cover each area of the target wafer and ensure that a sufficient number and quality of view point images are obtained, which are crucial for subsequent image stitching and marker point positioning.

[0066] Before moving the high magnification objective lens relative to the target wafer, the target wafer needs to be placed on the base of the three-axis motion platform, and a global coordinate system is established with the coordinate of the center of the target wafer in the ideal position (i.e. the position after accurate positioning) as the origin. The z-axis of the global coordinate system coincides with the rotation axis of the motion platform, and the plane in which the x-axis and y-axis of the global coordinate system lie is parallel to the target wafer on the base of the three-axis motion platform. When positioning the target wafer, the three-axis motion platform can rotate the target wafer along the z-axis, or translate the target wafer along the x-axis and y-axis to change the position coordinate of the target wafer.

[0067] wherein the spiral acquisition method refers to moving the high magnification objective lens above the target wafer in a spiral path in the clockwise or counterclockwise direction according to the preset moving step, and simultaneously intermittently photographing the target wafer (i.e. photographing once for each movement), thereby obtaining view point images covering different areas of the target wafer.

[0068] The preset marker point refers to a specific pattern or structure pre-set on the target wafer for assisting positioning, which can be a cross marker, a circular marker or a two-dimensional code, etc., and the ideal position and features thereof are known.

[0069] Specifically, the positioning module 2 performs the following when obtaining each view point image, performing image stitching on the adjacent acquired view point images in a nine-square grid stitching form with the currently obtained view point image as the center to obtain the stitched view point image, and obtaining at least two stitched images containing the preset marker point by performing marker point positioning on the stitched view point image through the template matching method. obtaining each view point image, performing image stitching on the adjacent acquired view point images in a nine-square grid stitching form with the currently obtained view point image as the center to obtain the stitched view point image, and obtaining at least two stitched images containing the preset marker point by performing marker point positioning on the stitched view point image through the template matching method. The template matching method is used to match each spliced view image with a preset marker point template image to detect whether any preset marker point identical to the preset marker point template image exists in each spliced view image. When it is detected that any preset marker point identical to the preset marker point template image exists in the current spliced view image, the current spliced view image is determined to be a spliced image containing the preset marker point, and the search for the current preset marker point is ended. The high-magnification objective lens is moved to an ideal position corresponding to another preset marker point in the target wafer, so that a spliced image containing the other preset marker point is obtained by using the same spiral acquisition method, image splicing method and template matching method, and at least two spliced images containing the preset marker point are obtained.

[0070] Specifically, the positioning module 2 acquires a view image, and then performs the following operations when a spliced view image is obtained by splicing adjacent acquired view images in a nine-square form with the acquired view image as the center: When a view image is acquired, adjacent images adjacent to the acquired view image in eight directions are extracted from the acquired view images according to the moving track of the high-magnification objective lens. The acquired view image is spliced with the adjacent images based on the positional relationship between the adjacent images and the acquired view image to obtain a spliced view image with the acquired view image as the center.

[0071] During wafer positioning, a single view image may not completely cover the preset marker point, so image splicing is needed to expand the field of view to ensure that the preset marker point can be completely captured. When a new view image is generated, images having a spatial adjacency relationship with the acquired view image in eight directions are selected from the acquired view images as adjacent images according to the moving direction and historical track of the high-magnification objective lens.

[0072] The acquired view image is spliced with the adjacent images based on the positional relationship between the adjacent images and the acquired view image to obtain a spliced view image with the acquired view image as the center.

[0073] In some optional embodiments, when the images are spliced, not only the nine-square splicing form can be used, but also other matrix arrangement methods such as QxQ or QxP (Q and P are positive integers) splicing forms can be used. The nearby acquired images are spliced with the current acquired viewpoint image as the center to obtain the spliced viewpoint image.

[0074] When the positioning module 2 is executed, whenever a new viewpoint image is acquired and spliced to obtain a spliced viewpoint image, template matching (for example, existing matching calculation methods such as a correlation coefficient method based on normalized cross-correlation, a contour shape-based recognition algorithm, a local feature point matching algorithm (such as SIFT or ORB), or a machine learning driven target detection method) is performed on the spliced viewpoint image and the preset marker point template image to detect the presence of the preset marker point in the spliced viewpoint image in real time. This real-time detection mechanism ensures that once the condition of the presence of the preset marker point is met during the image splicing process, the current spliced viewpoint image can be immediately determined as the target image, that is, the spliced image containing the preset marker point, and the search for the current preset marker point is ended, thereby avoiding unnecessary image acquisition and splicing and improving efficiency. The existing matching calculation methods are prior art and will not be described in detail here.

[0075] By the above method, only a high-magnification objective lens is used, images are sequentially acquired at multiple viewpoints around the ideal position of the preset marker point according to the multi-viewpoint image acquisition strategy, and the multi-viewpoint images are spliced and fused. The method of fusing "small images (viewpoint images)" into a large image (spliced image) enables the preset marker point to enter the field of view, fully utilizes the high spatial resolution of the high-magnification objective lens, and acquires high-precision positioning data of the preset marker point and the wafer. The method can simplify the hardware of the visual imaging system while maintaining high positioning accuracy and reduce the hardware cost of the positioning device.

[0076] Specifically, after the positioning module 2 performs marker point matching between each spliced viewpoint image and the preset marker point template image by the template matching method to detect whether any preset marker point identical to the preset marker point template image exists in each spliced viewpoint image, the positioning module 2 further performs: When the number of acquired viewpoint images is greater than or equal to the preset maximum viewpoint number threshold, and any preset marker point identical to the preset marker point template image has not been detected in each spliced viewpoint image, it is determined that the preset marker point search is timed out, the search for the preset marker point is terminated, and it is determined that the wafer positioning fails, The positioning module 2, when executed, counts the number of acquired viewpoint images in real time during image acquisition, and when the number of viewpoint images reaches a preset maximum viewpoint number threshold, and the preset marker point has not been detected, determines that the current wafer positioning process has exceeded the normal time range, terminates the marker point search process, and avoids invalid resource consumption.

[0077] In some optional embodiments, after a wafer positioning failure, another preset marker point can be selected for repositioning, and if positioning failures occur continuously, a worker is notified in time for abnormal handling.

[0078] The positioning module 2, when executed, adjusts the position of the high-magnification objective lens to the ideal position of another preset marker point immediately after detecting the first preset marker point, and quickly locks the position of another preset marker point (if the number of acquired viewpoint images is greater than or equal to the preset maximum viewpoint number threshold, another preset marker point can be selected for repositioning) by repeating the spiral acquisition method, image stitching method, and template matching method described above, to obtain at least two stitched images containing the preset marker points.

[0079] Specifically, when the computing module 3 calculates the angle offset of the target wafer according to the position coordinates of the preset marker points in the stitched images, it performs the following steps: According to the offset of the preset marker points from the viewpoint center positions of the corresponding stitched images in the stitched images, and in combination with a preset image spatial resolution, the position coordinates of the preset marker points are calculated. The position coordinates of any two preset marker points are input into a preset angle offset calculation formula, and the angle offset of the target wafer is calculated.

[0080] The computing module 3, when executed, determines the pixel offset of the preset marker point relative to the viewpoint center position according to the pixel number difference of the preset marker point from the viewpoint center position in the horizontal coordinate direction and the pixel number difference in the vertical coordinate direction in the stitched image in which the preset marker point is located. Subsequently, the pixel offset is multiplied by a preset image spatial resolution to calculate the offset coordinates of the preset marker point relative to the viewpoint center position. The offset coordinates and the coordinates of the viewpoint center position are combined to calculate the accurate position coordinates of the preset marker point on the wafer plane. That is, the position coordinates of the preset marker point can be calculated by the following marker point coordinate formula: ; ; wherein, is the horizontal coordinate of the preset marker point; is a horizontal coordinate of a preset mark point corresponding to a view center position of a spliced image; n is a pixel offset of the preset mark point relative to the view center position in a horizontal coordinate direction; and R is a preset image space resolution. is a vertical coordinate of the preset mark point; is a vertical coordinate of a preset mark point corresponding to a view center position of a spliced image; and m is a pixel offset of the preset mark point relative to the view center position in a vertical coordinate direction.

[0081] As shown in Figure 6 , Figure 6 is a schematic diagram of a preset mark point in a spliced image, wherein the circle a1, the circle a4, the circle a5, the circle a6, the circle a7, the circle a8, the circle a18, the circle a19, and the circle a20 are optical centers when a high magnification objective lens is moved each time, the solid line area p is a spliced image of a view area that has been collected, the circle a6 is a view center position of the spliced image, the dashed line rectangle c is a view area of an image that has not been collected (in nine-patch image splicing, if images of adjacent views have not been collected, all pixel points of the image corresponding to the view area are assigned a value of 0), d is a preset mark point, x1 is a horizontal coordinate of an offset coordinate of the preset mark point relative to the view center position, and y1 is a vertical coordinate of the offset coordinate of the preset mark point relative to the view center position. It can be known from Figure 6 that the horizontal coordinate x1 and the vertical coordinate y1 calculated by using the pixel offset and the image space resolution, in combination with actual coordinates of the view center position of the spliced image, can be used to calculate the horizontal coordinate and the vertical coordinate of the preset mark point d.

[0082] Based on the calculated position coordinates of the preset mark points, the position coordinates of any two preset mark points are input into a preset angle offset calculation formula to calculate an angle offset of the target wafer.

[0083] The preset angle offset calculation formula is specifically as follows: ; wherein, is an angle offset; is a two-parameter arctangent function, and an angle output range of the two-parameter arctangent function is ; is a horizontal coordinate of one of the preset mark points; is a vertical coordinate of the one of the preset mark points; is a horizontal coordinate of another of the preset mark points; is a vertical coordinate of the another of the preset mark points; is an included angle formed by a line connecting the two preset mark points (a line connecting one of the preset mark points and the another of the preset mark points) and a positive direction axis of a horizontal coordinate axis when the target wafer is located at an ideal position, that is, , a longitudinal coordinate of one of the preset mark points when in the ideal position, a longitudinal coordinate of one of the preset mark points when in the ideal position, a longitudinal coordinate of one of the preset mark points when in the ideal position, a longitudinal coordinate of one of the preset mark points when in the ideal position.

[0084] Specifically, the correction module 4 performs the following when performing angle correction on the target wafer based on the angle offset and calculating the position offset of the target wafer after angle correction to obtain the accurately positioned target wafer: performing angle correction on the target wafer based on the angle offset to obtain the target wafer after angle correction; calculating the position offset between the position coordinates of the preset mark points in the target wafer after angle correction and the preset target position coordinates to provide accurate position deviation compensation for subsequent detection or repair and obtain the accurately positioned target wafer.

[0085] When the correction module 4 is executed, the three-axis motion stage is rotated angle (counterclockwise rotation is positive and clockwise rotation is negative) relative to the vision imaging system using the previously calculated angle offset to adjust the rotation of the target wafer so that the overall angle of the target wafer is aligned with the ideal state (ideal position) to obtain the target wafer after angle correction, wherein the position coordinates of the preset mark points follow the rotation of the target wafer for angle correction, and the position coordinates of the preset mark points after angle correction are: ; ; wherein, is the longitudinal coordinate of the preset mark points after angle correction; is the longitudinal coordinate of the preset mark points before angle correction; is the longitudinal coordinate of the preset mark points after angle correction; is the longitudinal coordinate of the preset mark points before angle correction.

[0086] As Figure 7 shown, Figure 7 is a schematic diagram of the target wafer relative to the ideal wafer position, wherein, is the angle offset, the solid circle e is the target wafer, the dashed line e' is the ideal wafer position, d1 is one of the preset mark points, and d2 is the other preset mark point. As Figure 7 known, rotating the target wafer by angle can correct the angle of the target wafer e'.

[0087] After completing the angle correction, the position offset of the target wafer in the X and Y directions is calculated by comparing the actual position coordinates of the preset mark point after angle correction with its ideal position coordinates, and the target wafer is accurately positioned. For example, if the actual position of the preset mark point after angle correction is (x', y'), and the ideal position is (x t ,y t ), then the position offset is (Δx, Δy) = (x t -x', y t -y'). This position offset can provide precise offset compensation for subsequent inspection or repair, ensuring the accuracy of the inspection or repair. When the subsequent inspection equipment or repair equipment performs operations, it can fine-tune its own motion trajectory or action point based on this compensation, thereby ensuring that its operation can be accurately aligned with the target area on the wafer.

[0088] like Figure 8 As shown, Figure 8 Schematic diagram of the target wafer after angle correction relative to its ideal position, where the solid circle e is the target wafer, the dotted line e' is the ideal position of the wafer, d3 is one of the preset marking points, d4 is another preset marking point, d3' is the ideal position of one of the preset marking points, and d4' is the ideal position of the other preset marking point. Figure 8 It can be seen from FIG that the offset between the actual position coordinates of the preset marking point and its ideal position can be determined to determine the offset between the target wafer e and the ideal wafer position e′.

[0089] As can be known from the above, the wafer precise positioning device obtains the viewpoint images at different positions in the target wafer in sequence by adopting the spiral acquisition method of the high-magnification objective lens, the target wafer is provided with a plurality of preset mark points for positioning, after each viewpoint image is obtained, the obtained viewpoint image is spliced in the form of a nine-square grid, and the mark point positioning is performed on the spliced viewpoint image by the template matching method, at least two spliced images containing the preset mark points are obtained, the angle offset of the target wafer is calculated according to the position coordinates of the preset mark points in the spliced images, the angle correction is performed on the target wafer based on the angle offset, the position offset of the target wafer after the angle correction is calculated, and the precisely positioned target wafer is obtained. Therefore, the angle offset of the target wafer is calculated by using at least two preset mark points obtained by performing the image splicing and the mark point positioning on the viewpoint images obtained by the spiral acquisition method, the angle correction is performed on the target wafer, the position offset of the target wafer after the angle correction is calculated, and the precisely positioned target wafer is obtained. Only the spiral image acquisition is performed by using the high-magnification objective lens, on the one hand, the high resolution characteristic of the high-magnification objective lens is used to improve the positioning precision, and on the other hand, the multi-viewpoint image acquisition method of the spiral is used to expand the field of view range of the preset mark point search, and the objective lens does not need to be switched. The existing wafer precise positioning method solves the problems that the system is complex, the hardware cost is high, and the mechanical error is easy to trigger when the double-objective lens switching mode is used, and two objective lenses and the corresponding switching mechanism do not need to be configured and maintained, the hardware cost is significantly reduced, and the positioning efficiency of the wafer is improved.

[0090] Please refer to Figure 3 , Figure 3 A structural schematic diagram of an electronic device provided by an embodiment of the present application is provided, and the present application provides an electronic device, which comprises a processor 301 and a memory 302. The processor 301 and the memory 302 are interconnected and communicate with each other through a communication bus 303 and / or other forms of connection mechanism (not marked). The memory 302 stores a computer program executable by the processor 301. When the electronic device is running, the processor 301 executes the computer program to perform the wafer precise positioning method in any optional implementation manner of the above-mentioned embodiments to realize the following functions: the spiral acquisition method of the high-magnification objective lens is adopted to obtain viewpoint images at different positions in the target wafer in sequence, the target wafer is provided with a plurality of preset mark points for positioning, after each viewpoint image is obtained, the obtained viewpoint image is spliced in the form of a nine-square grid, and the mark point positioning is performed on the spliced viewpoint image by the template matching method, at least two spliced images containing the preset mark points are obtained, the angle offset of the target wafer is calculated according to the position coordinates of the preset mark points in the spliced images, the angle correction is performed on the target wafer based on the angle offset, the position offset of the target wafer after the angle correction is calculated, and the precisely positioned target wafer is obtained.

[0091] The computer readable storage medium provided by the embodiments of the present application stores a computer program, and the computer program is executed by a processor to implement the wafer precise positioning method in any optional implementation manner of the above embodiments to realize the following functions: a spiral acquisition method of a high magnification objective lens is adopted to sequentially acquire viewpoint images at different positions in a target wafer, a plurality of preset mark points for positioning are arranged in the target wafer, each time a viewpoint image is acquired, the acquired viewpoint image is spliced in a nine-square grid splicing form, and a mark point positioning is performed on the spliced viewpoint image by a template matching method, at least two spliced images containing the preset mark points are obtained, an angle offset of the target wafer is calculated according to position coordinates of the preset mark points in the spliced images, an angle correction is performed on the target wafer based on the angle offset, a position offset of the target wafer after the angle correction is calculated, and the precisely positioned target wafer is obtained. The storage medium can be realized by any type of volatile or non-volatile storage device or a combination thereof, such as a static random access memory (SRAM), an electrically erasable programmable read-only memory (EEPROM), an erasable programmable read-only memory (EPROM), a programmable read-only memory (PROM), a read-only memory (ROM), a magnetic memory, a flash memory, a magnetic disk or an optical disk.

[0092] In the embodiments provided by the present application, it should be understood that the disclosed device and method can be implemented in other ways. The device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There can be another division during actual implementation. For example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections between the units can be indirect couplings or communication connections through some communication interfaces, devices or units, and can be electrical, mechanical or in other forms.

[0093] In addition, the units described as separate components may or may not be physically separate, and the components displayed as units may or may not be physical units, and may be located in one place, or distributed on multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment scheme.

[0094] Furthermore, the functional modules in various embodiments of the present application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0095] In this paper, the relationship terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between the entities or operations.

[0096] The above is only an embodiment of the present application and is not used to limit the protection scope of the present application. For those skilled in the art, the present application can have various changes and variations. Any modification, equivalent replacement, improvement, etc. within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A wafer precise positioning method for positioning a wafer, characterized in that: Including steps: A spiral acquisition method using a high-magnification objective lens is used to sequentially acquire viewpoint images at different positions in a target wafer having a plurality of preset marking points for positioning. Each time a viewpoint image is acquired, the acquired viewpoint images are stitched together in a nine-square grid pattern, and marker points are located on the stitched viewpoint images using a template matching method to obtain at least two stitched images containing the preset marker points; Calculating the angular offset of the target wafer according to the position coordinates of the preset marking points in the stitched image; Based on the angle offset, the target wafer is angle-corrected, and the position offset of the target wafer after the angle correction is calculated to obtain a precisely positioned target wafer.

2. The wafer precise positioning method according to claim 1, characterized in that: Using a spiral acquisition method with a high-magnification objective lens, viewpoint images at different positions on the target wafer are acquired sequentially, including: Move the high-magnification objective lens to the ideal position corresponding to any preset mark point on the target wafer; A spiral acquisition method is adopted to control the high-magnification objective lens to move according to a preset moving step length, and the target wafer is photographed in sequence by the high-magnification objective lens to obtain viewpoint images at different positions in the target wafer.

3. The wafer precise positioning method according to claim 2, characterized in that: Each time a viewpoint image is acquired, the acquired viewpoint images are stitched in a nine-square grid format, and the marker points of the stitched viewpoint images are located by a template matching method to obtain at least two stitched images containing the preset marker points, including: Each time a viewpoint image is acquired, adjacent acquired viewpoint images are stitched together in a nine-square grid format, with the currently acquired viewpoint image as the center, to obtain a stitched viewpoint image. Performing marker point matching on each spliced ​​viewpoint image and a preset marker point template image by a template matching method to detect whether there is any preset marker point identical to the preset marker point template image in each spliced ​​viewpoint image; When it is detected that the current spliced ​​viewpoint image has any of the preset mark points that are identical to the preset mark point template image, the current spliced ​​viewpoint image is determined to be a spliced ​​image containing the preset mark point, and the search for the current preset mark point is ended. The high-magnification objective lens is moved to the ideal position corresponding to another preset mark point in the target wafer, so as to obtain a spliced ​​image containing another preset mark point through the same spiral acquisition method, image splicing method and template matching method, and obtain at least two spliced ​​images containing the preset mark points.

4. The wafer precise positioning method according to claim 3, characterized in that: Each time a viewpoint image is acquired, the adjacent acquired viewpoint images are stitched together in a nine-square grid format with the currently acquired viewpoint image as the center to obtain a stitched viewpoint image, including: Each time a viewpoint image is acquired, adjacent images adjacent to the currently acquired viewpoint image in eight directions are extracted from the acquired viewpoint images according to the movement trajectory of the high-magnification objective lens; Taking the currently acquired viewpoint image as the center and based on the positional relationship between the adjacent image and the currently acquired viewpoint image, the currently acquired viewpoint image and the adjacent image are spliced ​​to obtain a spliced ​​viewpoint image.

5. The wafer precise positioning method according to claim 3, characterized in that: After performing marker point matching on each spliced ​​viewpoint image and a preset marker point template image by a template matching method to detect whether any preset marker point identical to the preset marker point template image exists in each spliced ​​viewpoint image, the method further includes: When the number of acquired viewpoint images is greater than or equal to a preset maximum viewpoint number threshold, and no preset mark point identical to the preset mark point template image has been detected in each spliced ​​viewpoint image, it is determined that the preset mark point search has timed out, the search for the preset mark point is terminated, and the wafer positioning is determined to have failed.

6. The wafer precise positioning method according to claim 1, characterized in that: Calculating the angular offset of the target wafer according to the position coordinates of the preset marking point in the stitched image includes: Calculating the position coordinates of the preset marking point according to the offset between the preset marking point in the stitched image and the viewpoint center position of the corresponding stitched image, in combination with the preset image spatial resolution; The position coordinates of any two of the preset marking points are input into a preset angle offset calculation formula to calculate the angle offset of the target wafer.

7. The wafer precise positioning method according to claim 1, characterized in that: Based on the angle offset, the target wafer is angle-corrected, and the position offset of the target wafer after the angle correction is calculated to obtain a precisely positioned target wafer, including: Based on the angle offset, performing angle correction on the target wafer to obtain an angle-corrected target wafer; The position offset between the position coordinates of the preset mark point in the target wafer after angle correction and the preset target position coordinates is calculated to provide accurate position deviation compensation for subsequent inspection or repair, thereby obtaining a precisely positioned target wafer.

8. A wafer precision positioning device for positioning a wafer, characterized in that: include: An acquisition module, used to sequentially acquire viewpoint images at different positions in the target wafer using a spiral acquisition method using a high-magnification objective lens; The target wafer is provided with a plurality of preset marking points for positioning; a positioning module configured to stitch the acquired viewpoint images each time a viewpoint image is acquired, and to locate marker points on the stitched viewpoint images using a template matching method, to obtain at least two stitched images containing the preset marker points; A calculation module, configured to calculate an angular offset of the target wafer based on the position coordinates of the preset marking points in the stitched image; The correction module is used to perform angle correction on the target wafer based on the angle offset, and calculate the position offset of the target wafer after the angle correction to obtain a precisely positioned target wafer.

9. An electronic device, characterized in that: The method comprises a processor and a memory, wherein the memory stores a computer program executable by the processor, and when the processor executes the computer program, the method runs the steps of the wafer precise positioning method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the wafer precise positioning method according to any one of claims 1 to 7 are executed.

Images