A vision-based wafer pre-alignment system and method

Abstract

This invention relates to a vision-based wafer pre-alignment system and method. The vision-based wafer pre-alignment method includes: Step S1: Dividing the edge of the wafer into at least two parts circumferentially; Step S2: Controlling a rotation mechanism to drive the wafer to rotate, sequentially feeding each edge division into the camera's imaging range, allowing the camera to capture images of each edge division to obtain images of each edge division; Step S3: Processing the images of each edge division to obtain data of each edge division in the camera coordinate system; Step S4: Converting the data of each edge division in the camera coordinate system into data in the world coordinate system according to the relationship between the camera coordinate system and the world coordinate system; Step S5: Obtaining the edge data of the wafer based on the data of each edge division in the world coordinate system. According to the technical solution of this invention, wafer pre-alignment can be achieved.

Description

Technical Field

[0001] This invention relates to the field of wafer prealignment technology, and in particular to a vision-based wafer prealignment system and method. Background Technology

[0002] Wafer pre-alignment technology is crucial in semiconductor manufacturing, ensuring that wafers are initially aligned and positioned before entering the fine processing stages. Pre-alignment technology primarily utilizes advanced optical and vision systems to perform preliminary inspection and positioning of the wafer. By identifying characteristic markings on the wafer, such as notches or flat edges, the wafer's position and angle are determined. Pre-alignment technology is indispensable in semiconductor manufacturing; by improving alignment efficiency and accuracy, it ensures that each wafer is in optimal condition before entering critical processing steps, thereby improving the quality and reliability of the entire manufacturing process. Therefore, how to achieve wafer pre-alignment has become a problem that needs to be solved by those skilled in the art. Summary of the Invention

[0003] In view of this, the present invention provides a vision-based wafer pre-alignment system and method, the main technical problem to be solved being: how to achieve wafer pre-alignment.

[0004] To achieve the above objectives, the present invention mainly provides the following technical solutions:

[0005] Embodiments of the present invention provide a vision-based wafer pre-alignment system, which includes a camera and a rotation mechanism;

[0006] The rotating mechanism is used to drive the wafer to rotate, so that the edge of the wafer passes through the camera's imaging range in sequence.

[0007] The present invention also provides a vision-based wafer pre-alignment method for use in conjunction with the vision-based wafer pre-alignment system described above, the wafer pre-alignment method comprising:

[0008] Step S1: Divide the edge of the wafer into at least two parts in the circumferential direction;

[0009] Step S2: Control the rotating mechanism to drive the wafer to rotate, so that each edge division of the wafer is sequentially sent into the camera's imaging range, and the camera takes pictures of each edge division of the wafer to obtain images of each edge division of the wafer.

[0010] Step S3: Process the image of each edge segment to obtain the data of each edge segment in the camera coordinate system;

[0011] Step S4: Based on the relationship between the camera coordinate system and the world coordinate system, convert the data of each edge segment in the camera coordinate system into data in the world coordinate system;

[0012] Step S5: Obtain the edge data of the wafer based on the data of each edge division in the world coordinate system.

[0013] In some implementations, the image processing for each edge segmentation portion specifically includes:

[0014] Each edge segment is sequentially processed by filtering, histogram equalization, edge contour extraction and segmentation, and threshold segmentation.

[0015] In some embodiments, the wafer edge data includes the wafer's circular edge data;

[0016] And / or, when the wafer is characterized by a notch on its edge, the wafer edge data includes notch data;

[0017] And / or, when the wafer's feature markers are flat edges set on the edge, the wafer's edge data includes flat edge data.

[0018] In some embodiments, when the wafer edge data includes the wafer's circular edge data, the wafer pre-alignment method further includes:

[0019] Step S6: Obtain the wafer eccentricity information based on the wafer's edge data; the eccentricity information includes at least one of eccentricity angle and eccentricity distance;

[0020] Step S7: Based on the eccentricity information, determine whether the wafer needs to be corrected.

[0021] In some embodiments, when the wafer's feature markers are notches located on its edges, and the wafer's edge data includes notch data, the wafer pre-alignment method further includes:

[0022] Step S8: Obtain the location of the gap based on the gap data;

[0023] Step S9: Obtain the wafer ID position based on the position of the notch and the preset notch-wafer ID position relationship;

[0024] Step S10: Based on the wafer ID position, control the rotation mechanism to drive the wafer to rotate, so as to send the wafer ID into the camera's shooting range, so that the camera can take a picture of the wafer ID to obtain an image of the wafer ID;

[0025] Step S11: Identify the wafer ID based on the wafer ID image.

[0026] In some embodiments, when the wafer's feature markers are flat edges disposed on the edge, and the wafer's edge data includes flat edge data and rounded edge data, the wafer pre-alignment method further includes:

[0027] Step S12: Based on the flat edge data and the round edge data, obtain the distance between the flat edge of the wafer and the center of the wafer, as well as the perpendicular foot point corresponding to the flat edge.

[0028] By employing the above technical solutions, the vision-based wafer pre-alignment system and method of the present invention have at least the following beneficial effects:

[0029] 1. This invention enables pre-alignment of the wafer through steps such as camera calibration, wafer edge data acquisition, edge image data preprocessing, coordinate system transformation, and least squares fitting of circles.

[0030] 2. The wafer pre-alignment system of the present invention has a wafer ID recognition function, which can maintain good pre-alignment accuracy for warped or thin wafers; in addition, the camera can also take high-speed photos to improve detection efficiency.

[0031] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, the preferred embodiments of the present invention are described in detail below with reference to the accompanying drawings. Attached Figure Description

[0032] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0033] Figure 1 This is a schematic diagram of a vision-based wafer pre-alignment system provided in an embodiment of the present invention;

[0034] Figure 2 This is a flowchart of a vision-based wafer pre-alignment method provided in an embodiment of the present invention;

[0035] Figure 3 This is a schematic diagram of coordinate system transformation;

[0036] Figure 4 This is a schematic diagram of a flat side;

[0037] Figure 5 This is a schematic diagram of the gap;

[0038] Figure 6 This is a schematic diagram of edge fitting;

[0039] Figure 7 This is a schematic diagram of notch recognition.

[0040] Reference numerals: 1. Rotating mechanism; 2. Turntable; 3. Camera; 4. Coaxial light source; 5. Bar light source; 6. Surface light source; 100. Wafer. Detailed Implementation

[0041] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0042] It should be noted that if the embodiments of the present invention involve directional indicators (such as up, down, left, right, front, back, etc.), the directional indicators are only used to explain the relative positional relationship and movement of the components in a certain specific posture (as shown in the figure). If the specific posture changes, the directional indicators will also change accordingly.

[0043] Furthermore, if the embodiments of this invention involve descriptions such as "first" or "second," these descriptions are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined with "first" or "second" may explicitly or implicitly include at least one of those features. Additionally, the technical solutions of the various embodiments can be combined with each other, but this must be based on the ability of those skilled in the art to implement them. If the combination of technical solutions is contradictory or impossible to implement, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed by this invention.

[0044] like Figure 1 As shown, one embodiment of the present invention proposes a vision-based wafer pre-alignment system, which includes a camera 3 and a rotation mechanism 1. The rotation mechanism 1 is used to drive the wafer 100 to rotate, so that the edges of the wafer 100 sequentially pass through the imaging range of the camera 3.

[0045] In the above example, the rotating mechanism 1 can sequentially feed the edge of the wafer 100 into the imaging range of the camera 3, thereby enabling the camera 3 to take pictures of the edge of the wafer 100. Specifically, the camera 3 captures a portion of the wafer edge each time. After the wafer 100 rotates one full circle, the camera 3 can acquire images of various locations on the wafer edge through multiple shots, facilitating subsequent image processing to obtain the radius, angle information, and feature marker information of the wafer 100.

[0046] In some implementations, such as Figure 1As shown, the aforementioned vision-based wafer pre-alignment system may include a turntable 2 for mounting wafer 100. A rotation mechanism 1 drives the wafer 100 to rotate by driving the turntable 2. The rotation mechanism 1 may include a motor to drive the turntable 2 to rotate.

[0047] The aforementioned camera 3 can be a CCD camera, etc. The aforementioned vision-based wafer pre-alignment system may also include a light source, which is turned on when the camera 3 takes a picture. The light source may include a coaxial light source 4, a bar light source 5, and a surface light source 6. In a specific application example, the coaxial light source 4 and the bar light source 5 may both be located on the side of the wafer 100 closer to the camera 3, while the surface light source 6 may be located on the side of the wafer 100 away from the camera 3.

[0048] In some implementations, such as Figure 2-7 As shown, the present invention also provides a vision-based wafer pre-alignment method for use in conjunction with the vision-based wafer pre-alignment system described above. This wafer pre-alignment method includes:

[0049] Step S1: Divide the edge of wafer 100 into at least two parts in the circumferential direction.

[0050] The edge divisions of wafer 100 can work together to form the complete edge of wafer 100.

[0051] Step S2: Control the rotating mechanism 1 to drive the wafer 100 to rotate, so as to send each edge division of the wafer 100 into the imaging range of the camera 3 in sequence, so that the camera 3 can take pictures of each edge division of the wafer 100 to obtain images of each edge division of the wafer 100.

[0052] For example, the edge of wafer 100 can be divided into N parts circumferentially, where N is a positive integer greater than or equal to 2. Rotation mechanism 1 drives wafer 100 to rotate 360 ​​degrees / N each time, sequentially feeding the divided edge parts of wafer 100 into the imaging range of camera 3, allowing camera 3 to capture images of each edge division. The rotation angle of wafer 100 can be collected by an encoder; each time wafer 100 rotates 360 degrees / N, camera 3 is triggered to take an image of the corresponding edge division.

[0053] Step S3: Process the images of each edge segment obtained in step S2 to obtain the data of each edge segment in the camera coordinate system.

[0054] Specifically, the image processing for each edge segmentation can be performed as follows: filtering, histogram equalization, edge contour extraction and segmentation, and threshold segmentation for each edge segmentation.

[0055] Among them, filtering is used for noise reduction, histogram equalization is used for normalization, edge contour extraction and segmentation is used to separate circular data from marked features (such as gaps or flat edges), and threshold segmentation is used to find marked features, such as finding the centroid of the gap.

[0056] Step S4: Based on the relationship between the camera coordinate system and the world coordinate system, convert the data of each edge division part in the camera coordinate system into data in the world coordinate system.

[0057] The process of converting data from the camera coordinate system to data from the world coordinate system can be based on... Figure 3 To understand. Figure 3 In, Xc(x c y c Xw(x) is the camera coordinate system (the coordinates after camera calibration), with the preset center of camera 3 as the origin. w y w ) is the world coordinate system, with the rotation center of turntable 2 as the center point.

[0058] Camera coordinate system Xc(x c y c ) and world coordinate system Xw(x w y w The relationship between Xw and f can be expressed as follows: Xw = f + x c Yw = -(y c -h). f is the preset center of camera 3 and the rotation center of dial 2, both at x w The directional deviation, h, is the difference between the preset center of camera 3 and the rotation center of dial 2 in the y-direction. w Deviation in direction.

[0059] Using this formula, the data points Xnc(x) in the camera coordinate system nc y nc All of these can first be converted to Xnw(x) nw y nw Then, based on the wafer rotation angle θn corresponding to the camera image acquisition, the data point X'nw(x') in the world coordinate system can be converted using the following formula. nw y' nw ).in,

[0060] x' nw =x nw *cos(-θn)+y nw *sin(-θn);

[0061] y' nw =-x nw *sin(-θn)+y nw *con(-θn).

[0062] Where, x nw =f+x nc y nw =-(y nc -h).

[0063] Step S5: Obtain the edge data of wafer 100 based on the data of each edge division in the world coordinate system.

[0064] The edge data of the aforementioned wafer 100 may include the rounded edge data of the wafer 100. When the feature marker of the wafer 100 is a notch provided on the edge, the edge data of the wafer 100 may include notch data. When the feature marker of the wafer 100 is a flat edge provided on the edge, the edge data of the wafer 100 may include flat edge data.

[0065] In the above example, the edge data of wafer 100 can be used to achieve pre-alignment of wafer 100. For example, the characteristic marking data of wafer 100, such as notches or flat edges, can be used to compare with the preset position. If there is a deviation between the two, the wafer 100 can be driven to rotate by the rotating mechanism 1 to eliminate the deviation.

[0066] The technology of pre-aligning the wafer based on the edge data of wafer 100 is existing technology and will not be described in detail here.

[0067] In some embodiments, when the edge data of wafer 100 includes the rounded edge data of wafer 100, the wafer pre-alignment method further includes:

[0068] Step S6: Obtain the eccentricity information of wafer 100 based on the edge data of wafer 100. The eccentricity information includes at least one of eccentricity angle and eccentricity distance.

[0069] Step S7: Based on the eccentricity information, determine whether wafer 100 needs eccentricity correction. Specifically, if the eccentricity angle of wafer 100 is within a preset range, no adjustment is needed; if the eccentricity angle of wafer 100 is outside the preset range, adjustment is required. Similarly, if the eccentricity distance of wafer 100 is within a preset range, no adjustment is needed; if the eccentricity distance is outside the preset range, adjustment is required.

[0070] In some embodiments, when the feature marker of wafer 100 is a notch disposed on the edge, and the edge data of wafer 100 includes notch data, the aforementioned wafer pre-alignment method further includes:

[0071] Step S8: Obtain the location of the gap based on the gap data. The location information of the gap includes the location of the gap center.

[0072] Step S9: Obtain the wafer ID position based on the position of the notch and the preset notch-wafer ID position relationship.

[0073] The notch on wafer 100 and the wafer ID are relatively fixed in position and have a specific positional relationship. Once the position of the notch on wafer 100 is determined, the position of the wafer ID can be determined based on the notch-wafer ID positional relationship.

[0074] Step S10: Based on the wafer ID position, control the rotating mechanism 1 to drive the wafer 100 to rotate, so as to send the wafer ID into the shooting range of the camera 3, so that the camera 3 can take a picture of the wafer ID to obtain an image of the wafer ID.

[0075] Step S11: Identify the wafer ID based on the wafer ID image. The technique of identifying characters within an image is existing technology and will not be elaborated upon here.

[0076] This invention not only enables pre-alignment of wafer 100, but also has wafer ID recognition function, making it more comprehensive and practical.

[0077] In some embodiments, when the feature marker of wafer 100 is a flat edge disposed on the edge, and the edge data of wafer 100 includes flat edge data and rounded edge data, the wafer pre-alignment method further includes:

[0078] Step S12: Based on the flat edge data and round edge data, obtain the distance between the flat edge of wafer 100 and the center of the wafer, as well as the perpendicular point corresponding to the flat edge. Specifically, based on the flat edge data and round edge data, the flat edge and round edge of wafer 100 can be fitted using the least squares method, and then the center of the round edge and the center of the flat edge can be found. In this way, the distance between the center of the flat edge and the center of the round edge, as well as the perpendicular point corresponding to the flat edge, can be calculated.

[0079] It should be noted that the edge data of wafer 100 can be used to fit the edge data of wafer 100 using the least squares method, thereby calculating the radius of wafer 100 and the center position of wafer 100 in the world coordinate system.

[0080] The wafer pre-alignment system of this invention features wafer ID recognition and maintains good pre-alignment accuracy for warped or thin wafers 100. The camera 3 can perform high-speed photography, improving inspection efficiency.

[0081] The driver of the rotating mechanism 1 of the present invention can send a photo-taking signal to the camera 3. The rotating mechanism 1 drives the wafer 100 to rotate into position, which can trigger the camera 3 to acquire the wafer edge image. Through the steps of image preprocessing, edge extraction, coordinate system transformation, least squares fitting circle, etc., the center coordinates, angle information and wafer radius of the notch or flat edge are returned.

[0082] The wafer pre-alignment process of this invention requires steps including camera calibration, wafer edge data acquisition, edge image data preprocessing, coordinate system transformation, and least-squares circle fitting. Camera calibration requires distance calibration based on the camera's mounting position and the rotation center of the rotating mechanism 1; distortion correction and pixel equivalent calibration are performed based on the endo-extrinsic parameters of camera 3 calibrated using a calibration plate. Edge data acquisition requires the rotating mechanism 1 to rotate more than one revolution, and the camera 3 to acquire images at a specified angle through encoder position comparison. Edge image data preprocessing involves filtering (noise reduction), histogram equalization (normalization), edge contour extraction and segmentation (segmenting circular data and notch or flat edge data), and threshold segmentation (finding the centroid of the notch). Specifically: the segmented circular and flat edge data are transformed to the world coordinate system; the circular edge data undergoes least-squares circle fitting to obtain the wafer center and radius (calculating the eccentricity angle and eccentricity distance); the notch center is directly transformed to the world coordinate system to obtain the notch position information; the flat edge data is fitted with a least-squares line to find the distance between the line and the center of the circle, as well as its corresponding perpendicular point. The system can determine whether to perform offset correction based on the eccentricity information. Using the notch location information and the standard location of the wafer ID, the rotating mechanism 1 rotates to the wafer ID position, turns on the coaxial light and bar light, triggers the camera 3 to capture an image, and performs OCR recognition of the wafer ID.

[0083] The above description is merely a preferred embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural transformations made using the contents of the present invention's specification and drawings under the inventive concept of the present invention, or direct / indirect applications in other related technical fields, are included within the patent protection scope of the present invention.

Claims

1. A vision-based wafer pre-alignment method for use in conjunction with a vision-based wafer pre-alignment system, the vision-based wafer pre-alignment system comprising a camera (3) and a rotation mechanism (1); the rotation mechanism (1) is used to drive the wafer (100) to rotate, so that the edge of the wafer (100) sequentially passes through the imaging range of the camera (3); characterized in that, The wafer pre-alignment method includes: Step S1: Divide the edge of the wafer (100) into at least two parts in the circumferential direction; Step S2: Control the rotating mechanism (1) to drive the wafer (100) to rotate so that each edge division of the wafer (100) is sequentially sent into the imaging range of the camera (3) so that the camera (3) can take pictures of each edge division of the wafer (100) to obtain images of each edge division of the wafer (100). Step S3: Process the image of each edge segment to obtain the data of each edge segment in the camera coordinate system; Step S4: Based on the relationship between the camera coordinate system and the world coordinate system, convert the data of each edge segment in the camera coordinate system into data in the world coordinate system; Step S5: Obtain the edge data of the wafer (100) based on the data of each edge division in the world coordinate system; the edge data of the wafer (100) includes the circular edge data of the wafer (100); Step S6: Obtain the eccentricity information of the wafer (100) based on the edge data of the wafer (100); the eccentricity information includes at least one of the eccentricity angle and the eccentricity distance; Step S7: Based on the eccentricity information, determine whether the wafer (100) needs to be corrected.

2. The vision-based wafer pre-alignment method as described in claim 1, characterized in that, The image processing for each edge segmentation portion specifically involves: Each edge segment is sequentially processed by filtering, histogram equalization, edge contour extraction and segmentation, and threshold segmentation.

3. The vision-based wafer pre-alignment method as described in claim 1 or 2, characterized in that, When the feature marker of a wafer (100) is a notch set on the edge, the edge data of the wafer (100) includes the notch data; And / or, when the feature marker of the wafer (100) is a flat edge set on the edge, the edge data of the wafer (100) includes flat edge data.

4. The vision-based wafer pre-alignment method as described in claim 3, characterized in that, When the feature marker of the wafer (100) is a notch set on the edge, and the edge data of the wafer (100) includes notch data, the wafer pre-alignment method further includes: Step S8: Obtain the location of the gap based on the gap data; Step S9: Obtain the wafer ID position based on the position of the notch and the preset notch-wafer ID position relationship; Step S10: Based on the wafer ID position, control the rotation mechanism (1) to drive the wafer (100) to rotate so as to send the wafer ID into the shooting range of the camera (3) and let the camera (3) take a picture of the wafer ID to obtain an image of the wafer ID; Step S11: Identify the wafer ID based on the wafer ID image.

5. The vision-based wafer pre-alignment method as described in claim 3, characterized in that, When the feature marker of the wafer (100) is a flat edge set on the edge, and the edge data of the wafer (100) includes flat edge data and rounded edge data, the wafer pre-alignment method further includes: Step S12: Based on the flat edge data and the round edge data, obtain the distance between the flat edge of the wafer (100) and the center of the wafer, as well as the perpendicular foot point corresponding to the flat edge.

Images