Image processing method, system, device and storage medium

Abstract

The application discloses an image processing method, system, device and storage medium, wherein the image processing method comprises the following steps: providing a wafer to be measured, and a periodic pattern is formed on the surface of the wafer to be measured; acquiring corresponding wafer images of different regions on the surface of the wafer to be measured, wherein the wafer images comprise local wafer images; calculating image coordinates of the periodic pattern in each wafer image by using an image matching algorithm; and corresponding the coordinates of the periodic pattern on all the wafer images to the same image coordinate system, so as to splice all the wafer images into a complete wafer surface image. The image processing method provided by the application can realize high-precision local wafer image alignment, and avoids the problem that feature points used for splicing and alignment are prone to misalignment in the traditional method.

Description

Technical Field

[0001] The present invention relates to the technical field of optical detection equipment, and in particular to an image processing method, system, equipment and storage medium. Background Art

[0002] In the integrated circuit production process, it is often necessary to capture images of wafers for defect detection.

[0003] To accurately identify and locate defects and ensure inspection accuracy, higher requirements must be placed on details such as the size, shape, and brightness of the defect area. To achieve this, image acquisition resolution is generally required to reach 20 to 50 μm / pixel. However, some larger wafers cannot be fully captured with a single camera to obtain high-resolution images.

[0004] Traditional technology takes multiple images of different parts of the wafer under the same shooting conditions, then searches for local feature key points in the local wafer images, solves the transformation matrix to align the multiple local wafer images, and then processes the overlapping boundaries to stitch them into a complete wafer image.

[0005] However, since periodic patterns such as dies (core elements) are mostly formed during the actual wafer production process, when multiple local wafer images are stitched together into a complete wafer image, the overlapping area is used to find local feature key points and solve the transformation matrix method. Because the wafer surface has a periodic pattern, there will be periodic misalignment during the key point matching process; when the size of the overlapping boundary is smaller than the period of the pattern, it is impossible to accurately determine the local wafer image alignment. Summary of the Invention

[0006] The purpose of the present invention is to provide an image stitching method to solve the problem of easy misalignment in existing wafer image stitching methods.

[0007] To achieve the above object, the present invention provides an image processing method in a first aspect, comprising:

[0008] Providing a wafer to be tested, wherein a periodic pattern is formed on a surface of the wafer to be tested;

[0009] Acquiring corresponding wafer images at different areas on the surface of the wafer to be measured, wherein the wafer images include local wafer images;

[0010] Calculating the image coordinates of the periodic pattern in each of the wafer images using an image matching algorithm;

[0011] The coordinates of the periodic patterns on all the wafer images are mapped to the same image coordinate system, so as to splice all the wafer images into a complete wafer surface image.

[0012] Preferably, the step of mapping the coordinates of all the periodic patterns on the wafer images to the same coordinate system comprises:

[0013] Selecting a wafer image as a reference image, wherein the image coordinates of the reference image are a reference coordinate system;

[0014] Deducing the coordinates of the periodic pattern outside the reference image in the wafer region on the reference coordinate system based on the image coordinates, the minimum period size, the number of periods of the periodic pattern in each row and column, and the deflection angle of the periodic pattern arrangement in the reference image;

[0015] The coordinates of the periodic patterns of other wafer images are mapped to the reference coordinate system by using image coordinate space transformation.

[0016] Preferably, the image coordinate space transformation comprises: solving a perspective transformation matrix according to the coordinates of the periodic pattern deduced from the reference image on the reference coordinate system and the coordinates of the corresponding periodic pattern on the other wafer images.

[0017] Preferably, the image coordinate space transformation further includes:

[0018] According to the perspective transformation matrix, the coordinates of all periodic patterns on the other wafer images are transformed using image coordinate space transformation to obtain new coordinates thereof in the reference coordinate system.

[0019] Preferably, the physical environmental conditions for acquiring corresponding wafer images in different areas of the surface of the wafer to be measured include: a pitch angle between the camera and the object, an angle between the camera and the object, and / or a distance between the camera and the object.

[0020] Preferably, the wafer image includes a first local image and a second local image, the first local image is obtained by a left camera, and the second local image is obtained by a right camera, and the distance and pitch angle of the left camera and the right camera to the wafer to be measured are consistent.

[0021] Preferably, the step of stitching the first partial image and the second partial image into a complete wafer surface image comprises:

[0022] Set the image coordinates of the periodic pattern of the first partial image to P L (x, y), set the image coordinates of the periodic pattern of the second partial image to P R (x, y), where x ranges from (0, X1), X1 is the camera field of view width (in pixels), and the diameter R of the wafer to be measured is between one and two times the camera field of view width;

[0023] Use the image coordinate space transformation to transform the coordinate system P of the second local image R The center coordinates of all periodic patterns in (x, y) are transformed into the coordinate system P of the first partial image L (x, y), where the value range of the center coordinate x of the periodic pattern in the second partial image in the coordinate system of the first partial image is (X2, X3), where X2 <X1,X1<X3<2*X1。

[0024] Preferably, the step of stitching the first partial image and the second partial image into a complete wafer surface image further comprises:

[0025] The coordinate system P of the second partial image R The center coordinates of all periodic patterns in (x, y) are transformed into the coordinate system P of the first partial image L (x, y), the overlapping area between the first partial image and the second partial image is subjected to perspective change, wherein the overlapping area is: X = X1-(X2-X1)*0.5, and is within the range of K pixels on the left and right, that is, [XK, X+K];

[0026] In the stitched complete wafer surface image, a method for obtaining a new grayscale value of each pixel includes: obtaining the grayscale value of the corresponding area of ​​the first local image in the region [0, XK];

[0027] In the region [X+K, X3], the grayscale value of the region corresponding to the second local image is obtained;

[0028] In the [XK, X+K] region, the new grayscale value Among them, G l (x, y) is the grayscale value of the corresponding area of ​​the first local image, G r (x, y) is the grayscale value of the corresponding area of ​​the second partial image.

[0029] Preferably,

[0030] Calculating the coordinates of periodic patterns on different wafer images using an image matching algorithm includes: extracting periodic patterns on the wafer images;

[0031] Find the image coordinates (x,y) of the center of each complete periodic pattern.

[0032] Preferably, the step of finding the image coordinates of the center of each complete periodic pattern on the wafer image comprises:

[0033] Extracting the template image of the complete minimum periodic pattern;

[0034] Performing correlation operations on the wafer image from left to right and from top to bottom using the template image to obtain a matching coefficient matrix; wherein the parameter values ​​in the matching coefficient matrix are used to measure the similarity between the sub-regions of the template image in the wafer image;

[0035] The correlation coefficient is used for evaluation to determine each complete minimum period pattern on the wafer image.

[0036] Preferably, the image matching algorithm is template matching based on grayscale values, template matching based on gradient values, or phase correlation matching.

[0037] The present invention also provides an image processing system, comprising:

[0038] A wafer-to-be-tested providing unit, adapted to provide a wafer-to-be-tested wafer, wherein a periodic pattern is formed on a surface of the wafer-to-be-tested;

[0039] a wafer image acquisition unit, adapted to acquire at least two local wafer images of different areas on the surface of the wafer to be measured;

[0040] a coordinate calculation unit, adapted to calculate the coordinates of the periodic pattern in the local wafer image using an image matching algorithm;

[0041] The image stitching calculation unit is adapted to correspond the coordinates of the periodic patterns on all the partial wafer images to a same coordinate system, so as to stitch all the partial wafer images into a complete wafer surface image.

[0042] The present invention also provides an image processing method, comprising:

[0043] Providing an object to be tested, wherein a periodic pattern is formed on a surface of the object to be tested;

[0044] Acquiring corresponding surface images at different areas on the surface of the object to be tested, wherein the surface images include local surface images;

[0045] Calculating the image coordinates of the periodic pattern in each of the local surface images using an image matching algorithm;

[0046] The coordinates of the periodic patterns on all the local surface images are mapped to the same image coordinate system, so as to splice all the local surface images into a complete surface image of the object to be tested.

[0047] The present invention also provides an image processing system, comprising:

[0048] a test object providing unit, adapted to provide a test object, wherein a periodic pattern is formed on a surface of the test object;

[0049] A surface image acquisition unit, adapted to acquire corresponding surface images at different areas on the surface of the object to be tested, wherein the surface images include local surface images;

[0050] a coordinate calculation unit, adapted to calculate the image coordinates of the periodic pattern in each of the local surface images using an image matching algorithm;

[0051] The image stitching calculation unit is adapted to correspond the coordinates of the periodic patterns on all the local surface images to the same image coordinate system, so as to stitch all the local surface images into a complete surface image of the object to be measured.

[0052] The present invention also provides a device comprising at least one memory and at least one processor, wherein the memory stores one or more computer instructions, wherein the one or more computer instructions are executed by the processor to implement the image processing method as described above.

[0053] The present invention also provides a storage medium, wherein the storage medium stores one or more computer instructions, and the one or more computer instructions are used to implement the image processing method described above.

[0054] Due to the application of the above scheme, the present invention has the following advantages and effects compared with the prior art:

[0055] By using the technical solution of the present invention, the image coordinates of the periodic pattern on the wafer surface are mapped to the same image coordinate system to achieve the splicing of multiple local wafer images into a complete wafer surface image. This can achieve high-precision local wafer image alignment and avoid the problem of easy misalignment of feature points used for splicing and alignment in traditional methods.

[0056] Furthermore, using the technical solution of the present invention, the coordinates of the periodic patterns of other wafer images are mapped to the reference coordinate system using image coordinate space transformation, so that even if the overlapping area is smaller than the periodic pattern, all local wafer images can be matched and aligned on the same coordinate system to complete image stitching.

[0057] Furthermore, by using the technical solution of the present invention, the minimum period pattern on the wafer image surface is extracted by matching the minimum period template pattern, which can accurately identify each complete minimum period pattern on the wafer surface and has strong anti-interference ability. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Attachment Figure 1 A schematic diagram of an image processing method provided by an embodiment of the present invention;

[0059] Attachment Figure 2 A schematic diagram of a complete surface image of a wafer to be tested provided by an embodiment of the present invention;

[0060] Attachment Figure 3 To the attached Figure 4 Schematic diagram of different partial wafer images of a wafer to be tested provided by an embodiment of the present invention;

[0061] Attachment Figure 5 A schematic diagram of stitching different partial wafer images into a complete wafer image provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0062] In the image stitching method, this method uses the image coordinates of the periodic pattern in the local image for matching and alignment, avoiding the problem of easy misalignment when matching the feature points of the periodic pattern in traditional technology, and also avoiding the problem of the inability to perform coordinate transformation in image stitching in traditional technology due to the small overlapping area.

[0063] To make the above-mentioned objects, features, and advantages of the present application more clearly understood, the specific embodiments of the present application are described in detail below with reference to the accompanying drawings. The following description sets forth many specific details to facilitate a full understanding of the present application. However, the present application can be implemented in many other ways than those described herein, and those skilled in the art can make similar improvements without violating the scope of the present application. Therefore, the present application is not limited to the specific embodiments disclosed below.

[0064] In the present invention, unless otherwise specified or limited, the terms "installed," "connected," "connected," "fixed," etc. should be understood in a broad sense. For example, they can refer to fixed connection, detachable connection, or integration; they can refer to mechanical connection, direct connection, or indirect connection through an intermediate medium; they can refer to internal communication between two elements or interaction between two elements, unless otherwise specified. Those skilled in the art will understand the specific meanings of the above terms in this application based on specific circumstances.

[0065] In the present invention, the orientation or position relationship indicated by terms such as "center", "upper", "lower", "axial", "bottom", "inner" and "outer" is based on the orientation or position assembly relationship shown in the drawings. It is only for the convenience of describing the present application and simplifying the description, and does not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it should not be understood as a limitation on the present application.

[0066] Furthermore, the terms "first," "second," and the like are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of the technical features being referred to. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. Throughout the description of this application, "plurality" means at least two, such as two, three, etc., unless otherwise specifically defined.

[0067] Example 1

[0068] refer to Figure 1 As shown, an embodiment of the present invention provides a wafer image processing method, comprising:

[0069] Executing step S100: providing a wafer to be tested, wherein a periodic pattern is formed on a surface of the wafer to be tested;

[0070] Specifically, in this embodiment, the wafer to be tested can be a wafer processed by semiconductor process steps such as lithography, etching, and metallization. The material of the wafer can be semiconductor substrate materials such as single crystal silicon (Si), gallium arsenide (GaAs), gallium nitride (GaN), indium phosphide (InP), silicon carbide (SiC), and diamond (C). The periodic pattern on the surface of the wafer to be tested is a die (core element) pattern. In other embodiments, it can also be a periodically arranged characteristic pattern distributed on the entire surface of other wafers, which can be significantly identified in image recognition and has a standard consistent periodic distribution.

[0071] Specifically, refer to Figure 2 As shown in FIG, a complete ideal wafer surface pattern in this embodiment is formed, on which die (core element) patterns are distributed according to a periodic law, and each die pattern has a corresponding code die (i, j) on the wafer. Figure 2 In , i represents the row number where the die is located, j represents the column number where the die is located, i∈(-5,4), j∈(3,-4).

[0072] Executing step S200: acquiring corresponding wafer images at different areas on the surface of the wafer to be measured, wherein the wafer images include local wafer images;

[0073] Specifically, refer to Figures 3 and 4 In this embodiment, two local wafer images are obtained by left and right cameras. Figure 3 It is the first partial image F1 captured by the left camera. Figure 4 The second partial image F2 is captured by the right camera. During the capture process, the distance and pitch angle between the left camera and the right camera and the wafer to be measured are kept consistent.

[0074] In other embodiments, the same camera can be moved to different positions for shooting, or, after the same camera shoots the first local image, the wafer to be tested can be rotated or turned around in the same plane to allow the camera to shoot other local areas to obtain multiple wafer images of different areas on the surface of the wafer to be tested.

[0075] Alternatively, in another embodiment, a camera array consisting of three, four, or more cameras may be used to photograph the wafer to be tested, and the camera translation or wafer translation / rotation may also be combined.

[0076] In the above different shooting implementation methods, the distance and pitch angle between the camera and the wafer to be measured are always kept consistent in each shooting.

[0077] Furthermore, the first partial image F1 and the second partial image F2 overlap near their adjacent borders. Specifically, in this embodiment, the die portions near the left side of the first partial image F1 and the die portions near the right side of the second partial image F2 have overlapping labels. Furthermore, the dies in the overlapping region of the first partial image F1 and the second partial image F2 cover more than one column / row.

[0078] In other implementations, the partial wafer image may also be a plurality of images that can be stitched together to form a complete wafer image, and the stitching may be done by relative translation, rotation, scaling, projection, and other operations.

[0079] To facilitate explanation of the technical principles of the present invention, in this embodiment, the first partial image F1 and the second partial image F2 are partial images on the left and right sides of a complete wafer image, respectively, and the stitching method of the two is translation.

[0080] However, the description of the following specific embodiments cannot be understood as limiting all technical solutions and specific implementations that can be covered by the technical principles of the present invention.

[0081] Executing step S300: calculating the image coordinates of the periodic pattern in each of the wafer images using an image matching algorithm;

[0082] Specifically, the diagram is a schematic diagram of calculating the coordinates of periodic patterns on different wafer images using an image matching algorithm in this embodiment, which includes: extracting periodic patterns on the wafer image; and finding the image coordinates (x, y) of the center of each complete periodic pattern.

[0083] refer to Figure 3 and Figure 4 As shown, the template image A of the complete minimum periodic pattern is extracted;

[0084] Correlation operations are performed on the first partial image F1 and the second partial image F2 using a template image A having a complete periodic pattern from left to right along the X direction and from top to bottom along the Y direction to obtain a matching coefficient matrix. Parameter values ​​in the matching coefficient matrix are used to measure the degree of similarity between sub-regions of the template image in the wafer image. In this embodiment, the image matching algorithm is grayscale value-based template matching, gradient value-based template matching, or phase correlation matching.

[0085] The center coordinates of each complete minimum period pattern on the first partial image F1 and the second partial image F2 are then determined based on the parameter values ​​in the matching coefficient matrix.

[0086] Set the image coordinates of the periodic pattern of the first partial image to P L (x, y), set the image coordinates of the periodic pattern of the second partial image to P R (x, y). Wherein, the value range of x is (0, X1), X1 is the camera field of view width (pixel unit), and the diameter R of the wafer to be measured is between one and two times the camera field of view width. Specifically, in this embodiment, the images of the die on the adjacent sides of the first partial image F1 and the second partial image F2 are incomplete. Specifically, the periodic pattern die(1,3), die(1,2), die(1,1), die(1,0), die(1,-1), die(1,-2), die(1,-3), die(1,-4) in the first partial image F1 is an incomplete periodic pattern. The images of die(-2,3), die(-2,2), die(-2,1), die(-2,0), die(-2,-1), die(-2,-2), die(-2,-3), die(-2,-4) in the second partial image F2 are incomplete periodic patterns. The other dies are complete minimum periodic patterns.

[0087] After determining all the complete minimum period images, the center coordinates P of each complete minimum period pattern on the first partial image F1 in the image coordinate system are calculated on the image coordinate system of the first partial image F1 and the second partial image F2. L (x, y) and the center coordinate P in the image coordinate system of each complete minimum period pattern on the second partial image F2 R (x,y).

[0088] Executing step S400: mapping the coordinates of the periodic patterns on all the wafer images to the same image coordinate system, so as to splice all the wafer images into a complete wafer surface image.

[0089] In this embodiment, step S400 includes:

[0090] Step S410: selecting a wafer image as a reference image, where the image coordinates of the reference image are a reference coordinate system;

[0091] Step S420: Deducing the coordinates of the periodic pattern in the reference image, the minimum period size, the number of periods of the periodic pattern in each row and column, and the deflection angle of the periodic pattern arrangement, to obtain the coordinates of the periodic pattern in the reference coordinate system in the wafer region outside the reference image;

[0092] Step S430: using image coordinate space transformation to map the coordinates of other periodic patterns in the wafer image to the reference coordinate system.

[0093] Furthermore, the image coordinate space transformation is to solve a perspective transformation matrix based on the coordinates of the periodic pattern deduced from the reference image on the reference coordinate system and the coordinates of the corresponding periodic pattern on the other wafer images.

[0094] Furthermore, step S430 includes:

[0095] Step F31: Solving the perspective transformation matrix M based on the one-to-one correspondence between the coordinates of the periodic pattern in the wafer area outside the reference image obtained by expansion on the reference coordinate system and the coordinates of the corresponding periodic pattern on the corresponding local wafer image;

[0096] Step F32: according to the perspective transformation matrix M obtained by solving step F31, the coordinates of all periodic patterns on the corresponding local wafer image are converted into coordinates on the reference coordinate system using image coordinate space transformation.

[0097] The following combination Figures 3 to 5 The above steps are elaborated in detail according to the specific situation.

[0098] Execute step S410:

[0099] Selecting a wafer image as a reference image, wherein the image coordinates of the reference image are a reference coordinate system;

[0100] In this embodiment, reference Figure 3 As shown, the first partial image F1 is selected as the reference image, and the coordinate system where the first partial image F1 is located is the reference coordinate system.

[0101] Next, execute step S420:

[0102] Deducing based on the image coordinates, minimum period size, number of periods of the periodic pattern in each row and column, and deflection angle of the periodic pattern arrangement in the first partial image F1, coordinates of the periodic pattern outside the reference image in the wafer region on the reference coordinate system are obtained;

[0103] In this embodiment, the image coordinates of the periodic pattern of the first partial image F1 are P L (x,y), the center coordinate P of the adjacent periodic pattern in the X or Y direction L (x, y) calculates the average horizontal and vertical spacing Dx and Dy of the minimum period, which is the minimum period size in this embodiment.

[0104] Then, according to the image coordinates, minimum period size and number of periods of the periodic pattern in each row and column of the reference image, the coordinates P of the periodic pattern outside the reference image in the wafer area on the reference coordinate system are obtained. R '(x,y).

[0105] Specifically, in this embodiment, the center coordinates P of the die outside the first partial image F1 obtained by deducing the first partial image F1 are obtained through the above steps. R '(x,y).

[0106] Further, in this embodiment, referring to Figure 5 As shown in, according to the first local image F1, the center coordinates P of die(1,3), die(1,2), die(1,1), die(1,0), die(1,-1), die(1,-2), die(1,-3), die(1,-4), die(2,3), die(2,2), die(2,1), die(2,0), die(2,-1), die(2,-2), die(2,-3), die(2,-4), die(3,2), die(3,1), die(3,0), die(3,-1), die(3,-2), die(3,-3), die(4,1), die(4,0), die(4,-1), and die(4,-2) in the reference coordinate system are deduced. R '(x,y).

[0107] These dies do not have a complete period in the first partial image F1. However, these dies have center coordinates P in the image coordinate system of the second partial image F2. R (x,y).

[0108] Next, step S430 is performed: coordinates of other periodic patterns of the wafer image are mapped to the reference coordinate system using image coordinate space transformation.

[0109] Specifically, this step includes executing step F31: for the die listed above, according to their center coordinates P on the second partial image F2 R (x, y) and the center coordinate P calculated by the first partial image F1 R '(x,y), one-to-one correspondence in the two coordinate systems, using the relationship between spatial coordinate system transformations to solve the 3*3 perspective transformation matrix M.

[0110] Furthermore, the process of solving the 3*3 perspective transformation matrix M refers to the following expression:

[0111]

[0112] Among them, in the 3*3 perspective transformation matrix M,

[0113] The scaling ratio and rotation angle performed on the second partial image F2;

[0114] is the translation vector of the second partial image F2;

[0115] [c1 c2] is the projection vector of the second local image.

[0116] Next, the coordinate system P of the second partial image is transformed into R (x,y) is transformed into the coordinate system P of the first local image L (x,y).

[0117] Specifically, according to the above calculation, the perspective transformation matrix M is obtained, and the scaling ratio, rotation angle, translation vector and projection vector required for the second partial image F2 relative to the image coordinates of the first partial image F1 can be obtained.

[0118] Next, step F32 is performed: according to the perspective transformation matrix M, the center coordinate points of all dies in the second partial image F2 are transformed using the image coordinate space transformation to obtain new coordinates of all dies in the second partial image F2 in the reference coordinate system.

[0119] Then the first partial image F1 and the second partial image F2 are combined into one image, and the image is obtained as follows: Figure 5 An image of the complete wafer is shown.

[0120] Specifically, in the complete wafer image spliced ​​together, the value range of the horizontal axis x of the first partial image F1 is (0, X1), and the value range of the horizontal axis x of the second partial image F2 is (X2, X3).

[0121] Wherein, R∈(X1,2X1), R is the diameter of the wafer to be measured, X3-X2=X1, and the value of X1 is the camera field of view width M (pixel unit).

[0122] Among them, in the coordinate system of the first partial image, the value range of x in the second partial image is (X2, X3), where X2 <X1,X1<X3<2*X1。

[0123] The step of stitching the first partial image F1 and the second partial image F2 into a complete wafer surface image further includes:

[0124] The coordinate system P of the second partial image L The coordinates of the die of the second partial image F2 on (x, y) are converted into the coordinate system P of the first partial image R (x, y), the overlapping area between the first partial image F1 and the second partial image F2 is subjected to perspective change, wherein the overlapping area is: X = X1-(X2-X1)*0.5, and is within the range of K pixels on the left and right, that is, [XK, X+K];

[0125] Furthermore, according to the width of the perspective transformation matrix M and the first partial image F1, a perspective change may be performed on the overlapping area between the first partial image F1 and the second partial image F2.

[0126] In the stitched complete wafer surface image, the method for obtaining the new grayscale value of each pixel includes:

[0127] Take the grayscale value of the area corresponding to the first local image in the region [0, XK];

[0128] In the region [X+K, X3], the grayscale value of the region corresponding to the second local image is obtained;

[0129] In the [XK, X+K] region, the new grayscale value Among them, G l (x, y) is the grayscale value of the corresponding area of ​​the first local image, G r (x, y) is the grayscale value of the corresponding area of ​​the second partial image.

[0130] An embodiment of the present invention further provides a wafer image processing system, comprising:

[0131] A wafer-to-be-tested providing unit, adapted to provide a wafer-to-be-tested wafer, wherein a periodic pattern is formed on a surface of the wafer-to-be-tested;

[0132] a wafer image acquisition unit, adapted to acquire at least two local wafer images of different areas on the surface of the wafer to be measured;

[0133] a coordinate calculation unit, adapted to calculate the coordinates of the periodic pattern in the local wafer image using an image matching algorithm;

[0134] The image stitching calculation unit is adapted to correspond the coordinates of the periodic patterns on all the partial wafer images to a same coordinate system, so as to stitch all the partial wafer images into a complete wafer surface image.

[0135] Example 2

[0136] This embodiment also provides an image processing method, including:

[0137] Providing an object to be tested, wherein a periodic pattern is formed on a surface of the object to be tested;

[0138] Acquiring corresponding surface images at different areas on the surface of the object to be tested, wherein the wafer image includes a local surface image;

[0139] Calculating the image coordinates of the periodic pattern in each of the local surface images using an image matching algorithm;

[0140] The coordinates of the periodic patterns on all the local surface images are mapped to the same image coordinate system, so as to splice all the local surface images into a complete surface image of the object to be tested.

[0141] In addition, this embodiment further provides an image processing system, including:

[0142] a test object providing unit, adapted to provide a test object, wherein a periodic pattern is formed on a surface of the test object;

[0143] A surface image acquisition unit, adapted to acquire corresponding surface images at different areas on the surface of the object to be tested, wherein the wafer image includes a local surface image;

[0144] a coordinate calculation unit, adapted to calculate the image coordinates of the periodic pattern in each of the local surface images using an image matching algorithm;

[0145] The image stitching calculation unit is adapted to correspond the coordinates of the periodic patterns on all the local surface images to the same image coordinate system, so as to stitch all the local surface images into a complete surface image of the object to be measured.

[0146] In the above embodiment, the object to be tested can be the wafer to be tested in Example 1, or it can be any other product with a periodic pattern formed on its surface, such as a film layer or screen layer in the display manufacturing process. During calibration and actual use, the thickness difference between the object to be tested and the production product should be less than the camera's depth of field.

[0147] Example 3

[0148] Furthermore, an embodiment of the present invention also provides a device, which may be a personal computer (PC), a laptop computer, a server, or other device with data processing capabilities, including at least one memory and at least one processor, wherein the memory stores one or more computer instructions, wherein the one or more computer instructions are executed by the processor to implement all the image processing methods described above.

[0149] In an embodiment of the present invention, the device may include a processor (e.g., a central processing unit (CPU), a communication bus, a user interface, a network interface, and a memory. The communication bus is used to realize the connection and communication between these components; the user interface may include a display screen (Display) and an input unit such as a keyboard (Keyboard); the network interface may optionally include a standard wired interface, a wireless interface (such as a wireless fidelity WIreless-FIdelity, WI-FI interface); the memory may be a high-speed random access memory (RAM) or a stable memory (non-volatile memory), such as a disk memory. The memory may optionally be a storage device independent of the aforementioned processor 1001. Those skilled in the art will understand that the above hardware structure does not constitute a limitation of the present invention, and may include more or fewer components than shown in the figure, or a combination of certain components, or a different arrangement of components.

[0150] An embodiment of the present invention further provides a storage medium storing one or more computer instructions for implementing the wafer image processing method described above. Alternatively, the storage medium stores an image stitching program, wherein when the image stitching program is executed by a processor, the steps of the wafer image processing method described above are implemented.

[0151] The method implemented when the image stitching program is executed can refer to the various embodiments of the image processing method of the present invention, and will not be described in detail here.

[0152] The above embodiments are intended only to illustrate the technical concepts and features of the present invention. Their purpose is to enable those skilled in the art to understand the contents of the present invention and implement them accordingly. They are not intended to limit the scope of protection of the present invention. Any equivalent changes or modifications made in accordance with the spirit of the present invention are intended to be covered by the scope of protection of the present invention.

Claims

1. An image processing method, characterized in that: include: Providing a wafer to be tested, wherein a periodic pattern is formed on a surface of the wafer to be tested; Acquiring corresponding wafer images at different areas on the surface of the wafer to be measured, wherein the wafer images include local wafer images; Calculating the image coordinates of the periodic pattern in each of the wafer images using an image matching algorithm; Corresponding the coordinates of the periodic patterns on all the wafer images to the same image coordinate system to stitch all the wafer images into a complete wafer surface image; The step of mapping the coordinates of all the periodic patterns on the wafer images to the same image coordinate system includes: Selecting a wafer image as a reference image, wherein the image coordinates of the reference image are a reference coordinate system; Deducing the coordinates of the periodic pattern outside the reference image in the wafer region on the reference coordinate system based on the image coordinates, the minimum period size, the number of periods of the periodic pattern in each row and column, and the deflection angle of the periodic pattern arrangement in the reference image; Using image coordinate space transformation to map the coordinates of other periodic patterns of the wafer image to the reference coordinate system; The wafer image includes a first partial image and a second partial image, and the step of splicing the first partial image and the second partial image into a complete wafer surface image includes: Set the image coordinates of the periodic pattern of the first partial image to , set the image coordinates of the periodic pattern of the second partial image to ,in, The value range is , is the camera field of view width, the diameter of the wafer to be measured Between one and two times the width of the camera's field of view; Use image coordinate space transformation to transform the coordinate system of the second local image The center coordinates of all periodic patterns in the image are transformed into the coordinate system of the first partial image , to obtain a complete wafer surface image, wherein the center coordinates of the periodic pattern in the second partial image in the coordinate system of the first partial image are The value range is , where X2 <X1,X1<X3<2 X1.

2. The image processing method according to claim 1, wherein: The image coordinate space transformation includes: solving a perspective transformation matrix according to the coordinates of the periodic pattern deduced from the reference image on the reference coordinate system and the coordinates of the corresponding periodic pattern on the other wafer images.

3. The image processing method according to claim 2, wherein: The image coordinate space transformation further includes: According to the perspective transformation matrix, the coordinates of all periodic patterns on the other wafer images are transformed using image coordinate space transformation to obtain new coordinates thereof in the reference coordinate system.

4. The image processing method according to claim 1, wherein: The physical environment conditions for obtaining corresponding wafer images in different areas of the surface of the wafer to be measured include: The pitch angle between the camera and the object, the angle between the camera and the object, and / or the distance between the camera and the object.

5. The image processing method according to claim 1, wherein: The first partial image is obtained by a left camera, and the second partial image is obtained by a right camera. The distances and pitch angles of the left camera and the right camera to the wafer to be measured are consistent.

6. The image processing method according to claim 1, wherein: The coordinate system of the second local image is transformed by using the image coordinate space The center coordinates of all periodic patterns in the image are transformed into the coordinate system of the first partial image , to obtain a complete wafer surface image, including: Use image coordinate space transformation to transform the coordinate system of the second local image The center coordinates of all periodic patterns in the image are transformed into the coordinate system of the first partial image ; Performing a perspective change on an overlapping area between the first partial image and the second partial image, wherein the overlapping area is: , left and right pixels, that is, \left [ {XK,X+K} \right ] ; In the stitched complete wafer surface image, the method for obtaining the new grayscale value of each pixel includes: In \left [ {0,XK} \right ] The region takes the grayscale value of the region corresponding to the first local image; In \left [ {X+K,X3} \right ] The grayscale value of the area corresponding to the second local image is taken as the area; In \left [ {XK,X+K} \right ] area, the new grayscale value ,in, is the grayscale value of the area corresponding to the first local image, is the grayscale value of the area corresponding to the second partial image.

7. The image processing method according to claim 1, wherein: Calculating the coordinates of periodic patterns on different wafer images using an image matching algorithm includes: extracting periodic patterns on the wafer images; Find the image coordinates of the center of each complete periodic pattern .

8. The image processing method according to claim 6, wherein: The step of finding the image coordinates of the center of each complete periodic pattern on the wafer image comprises: Extracting the template image of the complete minimum periodic pattern; Performing correlation operations on the wafer image from left to right and from top to bottom using the template image to obtain a matching coefficient matrix; wherein the parameter values ​​in the matching coefficient matrix are used to measure the similarity between the sub-regions of the template image in the wafer image; The correlation coefficient is used for evaluation to determine each complete minimum period pattern on the wafer image.

9. The image processing method according to claim 1, wherein: The image matching algorithm is template matching based on grayscale values, template matching based on gradient values, or phase correlation matching.

10. An image processing system, characterized in that: include: A wafer-to-be-tested providing unit, adapted to provide a wafer-to-be-tested wafer, wherein a periodic pattern is formed on a surface of the wafer-to-be-tested; a wafer image acquisition unit, adapted to acquire at least two local wafer images of different areas on the surface of the wafer to be measured; a coordinate calculation unit, adapted to calculate the coordinates of the periodic pattern in the local wafer image using an image matching algorithm; An image stitching calculation unit is adapted to map the coordinates of the periodic patterns on all the partial wafer images to the same image coordinate system so as to stitch all the wafer images into a complete wafer surface image, comprising: A wafer image is selected as a reference image, and the image coordinates of the reference image are a reference coordinate system; the image coordinates of the periodic pattern in the reference image, the minimum period size, the number of periods of the periodic pattern in each row and column, and the deflection angle of the periodic pattern arrangement are deduced to obtain the coordinates of the periodic pattern outside the reference image in the wafer area on the reference coordinate system; the coordinates of the periodic pattern of other wafer images are mapped to the reference coordinate system by using image coordinate space transformation; the wafer image includes a first local image and a second local image, and the image coordinates of the periodic pattern of the first local image are set to , set the image coordinates of the periodic pattern of the second partial image to ,in, The value range is , is the camera field of view width, the diameter of the wafer to be measured Between one and two times the camera field of view width; use the image coordinate space transformation to transform the coordinate system of the second local image The center coordinates of all periodic patterns in the image are transformed into the coordinate system of the first partial image , to obtain a complete wafer surface image, wherein the center coordinates of the periodic pattern in the second partial image in the coordinate system of the first partial image are The value range is , where X2 <X1,X1<X3<2 X1.

11. An image processing method, characterized in that: include: Providing an object to be tested, wherein a periodic pattern is formed on a surface of the object to be tested; Acquiring corresponding surface images at different areas on the surface of the object to be tested, wherein the surface images include local surface images; Calculating the image coordinates of the periodic pattern in each of the local surface images using an image matching algorithm; The coordinates of the periodic patterns on all the local surface images are mapped to the same image coordinate system to stitch all the local surface images into a complete surface image of the object to be tested, including: selecting a wafer image as a reference image, the image coordinates of the reference image being the reference coordinate system; deducing based on the image coordinates of the periodic pattern in the reference image, the minimum period size, the number of periods of the periodic pattern in each row and column, and the deflection angle of the periodic pattern arrangement, to obtain the coordinates of the periodic pattern outside the reference image in the wafer area on the reference coordinate system; using image coordinate space transformation to map the coordinates of the periodic patterns of other wafer images to the reference coordinate system; the image of the object to be tested includes a first local image and a second local image, and the image coordinates of the periodic pattern of the first local image are set to , set the image coordinates of the periodic pattern of the second partial image to ,in, The value range is , is the camera field of view width, the diameter of the wafer to be measured Between one and two times the camera field of view width; use the image coordinate space transformation to transform the coordinate system of the second local image The center coordinates of all periodic patterns in the image are transformed into the coordinate system of the first partial image , to obtain a complete surface image of the object to be tested, wherein the center coordinates of the periodic pattern in the second partial image in the coordinate system of the first partial image are The value range is , where X2 <X1,X1<X3<2 X1.

12. An image processing system, characterized in that: include: a test object providing unit, adapted to provide a test object, wherein a periodic pattern is formed on a surface of the test object; A surface image acquisition unit, adapted to acquire corresponding surface images at different areas on the surface of the object to be tested, wherein the surface images include local surface images; a coordinate calculation unit, adapted to calculate the image coordinates of the periodic pattern in each of the local surface images using an image matching algorithm; The image stitching calculation unit is suitable for mapping the coordinates of the periodic patterns on all the local surface images to the same image coordinate system so as to stitch all the local surface images into a complete surface image of the object to be tested, including: selecting a wafer image as a reference image, the image coordinates of the reference image being the reference coordinate system; deducing according to the image coordinates of the periodic pattern in the reference image, the minimum period size, the number of periods of the periodic pattern in each row and column, and the deflection angle of the periodic pattern arrangement, to obtain the coordinates of the periodic pattern outside the reference image in the wafer area on the reference coordinate system; using image coordinate space transformation to map the coordinates of the periodic patterns of other wafer images to the reference coordinate system; the image of the object to be tested includes a first local image and a second local image, and the image coordinates of the periodic pattern of the first local image are set to , set the image coordinates of the periodic pattern of the second partial image to ,in, The value range is , is the camera field of view width, the diameter of the wafer to be measured Between one and two times the camera field of view width; use the image coordinate space transformation to transform the coordinate system of the second local image The center coordinates of all periodic patterns in the image are transformed into the coordinate system of the first partial image , to obtain a complete surface image of the object to be tested, wherein the center coordinates of the periodic pattern in the second partial image in the coordinate system of the first partial image are The value range is , where X2 <X1,X1<X3<2×X1。 13. An image processing device, characterized in that The method comprises at least one memory and at least one processor, wherein the memory stores one or more computer instructions, wherein the one or more computer instructions are executed by the processor to implement the image processing method according to any one of claims 1 to 9 or claim 11.

14. A storage medium, characterized in that The storage medium stores one or more computer instructions, and the one or more computer instructions are used to implement the image processing method according to any one of claims 1 to 9 or claim 11.

Images