Crystal grain positioning method and device, electronic equipment and storage medium

Abstract

This specification provides a method, apparatus, electronic device, and storage medium for locating a die. The method includes: acquiring a wafer image; determining the matching coordinates of each potential object in the wafer image through feature matching; identifying potential objects located along a first axis and a second axis; and determining the periodic size of the die along the corresponding axis based on the spacing distribution between the matching coordinates of the potential objects along each axis, wherein the first axis and the second axis are perpendicular to each other; generating the theoretical coordinates of each die based on the periodic size of the die and the position index of each die; and determining the actual coordinates of each die based on the theoretical coordinates of each die and the matching coordinates of potential objects within its vicinity.

Description

Technical Field

[0001] This specification relates to the field of semiconductor technology, and in particular to methods, apparatus, electronic devices and storage media for positioning chips. Background Technology

[0002] A wafer is the basic substrate material for semiconductor manufacturing, on which a large number of repeating integrated circuit structural units are formed. These structural units, which have not been physically separated after all the processes are completed, are collectively called dies. Each die is a fully functional integrated circuit die, possessing specific electrical properties and logic functions.

[0003] In the metrology and inspection process, precise positioning of the dies on the wafer is the foundation and prerequisite for all subsequent processes. By establishing a precise die coordinate system, metrology and inspection equipment can accurately align with the area to be measured. The accuracy and stability of the positioning directly determine the repeatability and consistency of the measurement results, and are the core support for ensuring closed-loop process control, optimizing yield, and reducing production costs. Summary of the Invention

[0004] To overcome the problems existing in related technologies, this specification provides methods, apparatus, electronic devices and storage media for positioning grains.

[0005] According to a first aspect of the embodiments of this specification, a method for positioning grains is provided, the method comprising: A wafer image is acquired, and the matching coordinates of each potential object in the wafer image are determined by feature matching.

[0006] Potential objects located on the first axis and the second axis are identified. Based on the spacing distribution between the matching coordinates of the potential objects on each axis, the periodic size of the grain on the corresponding axis is determined. The first axis and the second axis are perpendicular to each other.

[0007] The theoretical coordinates of each grain are generated based on the periodic size of the grain and the position index of each grain.

[0008] The actual coordinates of each grain are determined based on its theoretical coordinates and the matching coordinates of potential objects in its vicinity.

[0009] According to a second aspect of the embodiments of this specification, a grain positioning device is provided, comprising: The matching coordinate determination module is used to acquire a wafer image and determine the matching coordinates of each potential object in the wafer image through feature matching.

[0010] The period size determination module is used to determine potential objects located on the first axis and the second axis, and to determine the period size of the grain on the corresponding axis based on the spacing distribution between the matching coordinates of the potential objects on each axis; the first axis and the second axis are perpendicular to each other.

[0011] The theoretical coordinate determination module is used to generate the theoretical coordinates of each grain based on the period size and the position index of each grain.

[0012] The actual coordinate determination module is used to determine the actual coordinates of each grain based on its theoretical coordinates and the matching coordinates of potential objects in its vicinity.

[0013] According to a third aspect of the embodiments of this specification, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the method as described in the first aspect.

[0014] According to a fourth aspect of the embodiments of this specification, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the method as described in the first aspect.

[0015] The technical solutions provided in the embodiments of this specification may include the following beneficial effects: In the embodiments of this specification, firstly, when generating the theoretical coordinates of the grains, this method does not rely on the nominal periodic size of the grains. Instead, it derives the actual periodic size of the grains based on the spacing distribution between the matching coordinates of potential objects extracted along the mutually perpendicular first and second axes in the wafer image. Then, based on the actual periodic size and the position index of each grain, the theoretical coordinates of the grains are generated so that the generated theoretical coordinates can reflect the current operating conditions of the wafer. It can be compatible with the distribution of different grains of the same wafer type; it can also be compatible with wafers of different types and processes, and can be applied to uncut wafers and wafer expansion / dicing.

[0016] Secondly, regardless of whether the die is successfully matched, the corresponding actual coordinates can be generated based on the theoretical coordinates of the die and the matching coordinates of potential objects in its vicinity. This makes up for the location of dies that cannot be identified by feature matching. It can be used for wafers with large defects, color differences, or a large number of missing dies, and is compatible with the location of MPW (Multi-Project Wafer) type wafers.

[0017] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this specification. Attached Figure Description

[0018] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this specification and, together with the description, serve to explain the principles of this specification.

[0019] Figure 1 This is a flowchart illustrating a method for positioning a grain according to an exemplary embodiment of this specification.

[0020] Figure 2 This is a schematic diagram of an image of a complete wafer illustrated according to an exemplary embodiment of this specification.

[0021] Figure 3 This is a schematic diagram of an image of a local wafer region shown in this specification according to an exemplary embodiment.

[0022] Figure 4 This is a schematic diagram of a feature template image of a grain shown in this specification according to an exemplary embodiment.

[0023] Figure 5 This is a schematic diagram illustrating a candidate potential object within a preset range in a first and second axis according to an exemplary embodiment of this specification.

[0024] Figure 6 This is a schematic diagram of all potential objects in the positive direction of the first axis according to an exemplary embodiment of this specification.

[0025] Figure 7 This is a schematic diagram of all potential objects in the positive and negative directions of a first axis, as illustrated in this specification according to an exemplary embodiment.

[0026] Figure 8 This is a schematic diagram illustrating the theoretical coordinates of a grain determined based on the period size and the grain position index, according to an exemplary embodiment of this specification.

[0027] Figure 9 This is a schematic diagram illustrating a region of grain division according to an exemplary embodiment of this specification.

[0028] Figure 10 This is a schematic diagram illustrating a rectangular shape in the vicinity of the theoretical coordinates of a grain according to an exemplary embodiment of this specification.

[0029] Figure 11 This is a schematic diagram of the structure of an electronic device according to an exemplary embodiment of this specification.

[0030] Figure 12 This is a block diagram illustrating a die positioning device according to an exemplary embodiment of this specification. Detailed Implementation

[0031] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this specification. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this specification as detailed in the appended claims.

[0032] The terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to be limiting of this specification. The singular forms “a,” “the,” and “the” as used in this specification and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any and all possible combinations of one or more of the associated listed items.

[0033] It should be understood that although the terms first, second, third, etc., may be used in this specification to describe various information, this information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of this specification, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to determination."

[0034] A wafer is the basic substrate material for semiconductor manufacturing, on which a large number of repeating integrated circuit structural units are formed. These structural units, which have not been physically separated after all the processes are completed, are collectively called dies. Each die is a fully functional integrated circuit die, possessing specific electrical properties and logic functions.

[0035] In the metrology and inspection process, precise positioning of the dies on the wafer is the foundation and prerequisite for all subsequent processes. By establishing a precise die coordinate system, metrology and inspection equipment can accurately align with the area to be measured. The accuracy and stability of the positioning directly determine the repeatability and consistency of the measurement results, and are the core support for ensuring closed-loop process control, optimizing yield, and reducing production costs.

[0036] Existing die positioning technologies are mostly based on feature matching algorithms, which determine the die coordinates by matching features between a pre-set die template image and a wafer image. However, color difference fluctuations, random defects or contamination on the wafer surface that occur during semiconductor processing often lead to feature recognition failure, resulting in misalignment or missed points.

[0037] Ideally, grain coordinates can be determined by a fixed periodic dimension and position index. However, in processes involving physical deformation, such as wafer expansion and dicing, grain distribution exhibits high nonlinear instability. The physical span between grains no longer maintains its initial consistency but fluctuates with changes in spatial position. Taking the wafer expansion process as an example, due to tensile stress, the grain spacing from the center region to the edge region of the wafer increases significantly, breaking the original equidistant arrangement. Therefore, calculating the coordinates of each grain solely based on its nominal periodic dimension and position index is unsuitable for scenarios where wafer operating conditions change.

[0038] To address the aforementioned technical issues, this specification provides a method for positioning the die, which can not only handle the positioning of standard, perfect wafers, but also effectively cope with complex working conditions such as stretching, deformation, damage, or heterogeneous arrangement.

[0039] The embodiments described in this specification will now be described in detail.

[0040] Figure 1 This is a flowchart illustrating a method for positioning a grain according to an exemplary embodiment of this specification, such as... Figure 1 As shown, the method for locating this grain includes the following steps: Step 101: Obtain the wafer image and determine the matching coordinates of each potential object in the wafer image through feature matching.

[0041] Wafer images can be images obtained by taking pictures of the wafer surface using image acquisition devices (such as high-magnification industrial cameras, electron scanning microscopes, etc.).

[0042] like Figure 2 As shown, a wafer image can be a complete image covering the entire wafer size acquired in one go; or it can be a partial scan of the wafer surface row by row or column by column, acquiring multiple partial field-of-view images of different parts of the wafer (e.g., such as...). Figure 3 The partial view shown is one example, and the images are then stitched together to obtain the complete image. This specification does not limit the method of generating wafer images.

[0043] Feature matching can be achieved by translating the grain template image onto the wafer image, searching for potential objects in the wafer image that are similar to the grain template image, and recording the matching coordinates of these potential objects.

[0044] A grain template image can be a complete image of the grain or an image of a local region within the grain that possesses unique characteristics. For example, such as... Figure 4The grain template image shown selects the two circles in the middle row of the grain and the area between them as the grain template. These two circles correspond to the key functional areas on the grain surface, such as metal bumps and alignment marks. The area between them contains the surrounding physical texture. This part has unique features and can be used as a grain template. Figure 4 The die template image can be obtained by the operator selecting a portion of the wafer area and then using the image of that portion as the die template image.

[0045] The translation method can be from left to right or from top to bottom, with a preset number of pixels at each step, translating the die template image onto the wafer image. For example, the preset number of pixels could be 2 or 3 pixels, etc.

[0046] After each translation, the similarity between the grain template image and the translated wafer region can be compared. For example, the pixels of the grain template image and the corresponding wafer region can be converted into vector representations, and the vector similarity between the two can be compared. For example, the edge contours (e.g., corner points) of the wafer region can be compared to see if they overlap with the grain template image, and the similarity between the two can be determined based on the overlap rate.

[0047] After identifying a wafer region similar to the grain template image, this wafer region can be considered a potential object. A potential object can refer to an object that is suspected to be a grain but has not been verified during the feature matching process. For example, a potential object could be a matched grain or a matched non-grain object, such as interference on the wafer surface.

[0048] Step 102: Identify potential objects located on the first axis and the second axis, and determine the periodic size of the grain on the corresponding axis based on the spacing distribution between the matching coordinates of the potential objects on each axis; the first axis and the second axis are perpendicular to each other.

[0049] The first axis and the second axis can be the "row direction" and "column direction" that correspond to the arrangement of grains on the wafer, respectively.

[0050] In an exemplary implementation of determining potential objects located along the first and second axes: In one embodiment, the axes of a first axis and a second axis can be determined first, and then potential objects along the first and second axes can be identified. For example, the center of the wafer can be used as the origin, the horizontal direction of the image can be used as the first axis, and the vertical direction as the second axis. Then, specific tolerance ranges on both sides of the first and second axes are determined, and potential objects in the wafer image whose matching coordinates fall within the specific tolerance range on both sides of the axis are classified as potential objects along that axis. For example, the tolerance range can be the range between parallel lines with a preset width as the center line. The preset width can be set to the size of one half-die.

[0051] In another embodiment, a starting search point can be determined, and using the starting search point as a seed, search points that satisfy the spatial proximity condition can be identified step by step along the first axis and the second axis, respectively, as potential objects on the corresponding axis.

[0052] For example, a starting search point can be determined, and using the starting search point as a seed, on the first axis, search points that satisfy the spatial proximity condition on the first axis can be identified step by step as potential objects on the first axis based on spatial proximity; on the second axis, search points that satisfy the spatial proximity condition on the second axis can be identified step by step as potential objects on the second axis based on spatial proximity.

[0053] Spatial proximity is used to evaluate the consistency of the spatial distribution between potential objects and the current search point. Specifically, this condition can be defined as the physical distance between two points being within a preset threshold, or their relative orientation being within a specified angular deviation range.

[0054] During execution, the initial search point is used as a seed, and potential objects that meet the requirements are determined by spatial proximity conditions as search points for the next round of iteration. This process is repeated iteratively until all search points that meet the spatial proximity conditions along the axis are found as potential objects along the corresponding axis.

[0055] In this embodiment, although the film expansion causes a large absolute displacement of the wafer edge relative to the center, the relative displacement deviation between two adjacent grains is very small. By gradually determining potential objects through the above iterative method, the search direction for each round can be determined based on the position of the search point in that round, allowing for fine-tuning of the search direction for the next iteration. Even if the grains shift, the search direction is determined based on the shifted grains. Compared to the method of first determining the axis and then determining the potential objects in the vicinity of the axis, the method in this embodiment is more adaptable to wafer variations.

[0056] When determining the starting search point, potential objects located near the center of the wafer can be selected as candidate potential objects. If a candidate potential object meets the matching criteria, it is then determined as the starting search point.

[0057] Matching conditions may include: the candidate potential object exists simultaneously with at least one other potential object within a preset range of the first axis and the second axis.

[0058] For example, a potential object can be randomly selected from the vicinity of the center of the wafer as a candidate potential object. If the candidate potential object meets the matching criteria, it is determined as the starting search point. If the selected potential object does not meet the matching criteria, other nearby potential objects can be selected and their matching criteria can be checked again until a potential object that meets the matching criteria is found as the starting search point.

[0059] In this embodiment, by determining whether at least one other potential object exists simultaneously within a preset range of the first and second axes, it can be inferred whether the candidate potential object is a matched grain rather than a matching failure point, thereby ensuring the reliability of the seed used for iterative search. A matching failure point can refer to a matched non-grain object, such as interference, defects, or stains on the wafer surface.

[0060] In one embodiment, such as Figure 5 As shown, the preset range can be set with the candidate potential object as the center, and generate a rectangle with a length of N grain sizes and a width equal to the maximum offset between adjacent grains at the center of the wafer, respectively, along the first axis and the second axis.

[0061] The value of N can be no less than 3, and the maximum offset can be set to one-fifth or one-quarter of the grain size, but not exceeding half the grain size.

[0062] If at least one other potential object (besides the candidate potential object) exists within the first rectangle, and at least one other potential object (besides the candidate potential object) also exists within the second rectangle, then the candidate potential object meets the matching criteria and is a successfully matched grain. If at least one other potential object exists within only one of the two rectangles, while no other potential object exists within the other rectangle, then the candidate potential object does not meet the matching criteria and is a matching failure point.

[0063] In this embodiment, it is difficult for other successfully matched grains to exist simultaneously within a preset range in both the horizontal and vertical directions at the matching failure point. By generating two orthogonal rectangular ranges centered on the candidate potential object, matching failure points in the first axis and the second axis can be excluded at the same time, thereby ensuring that the starting search point is the actual grain matched by the feature rather than the interference item of the matching failure.

[0064] In practical applications, multiple candidate potential objects that meet the matching criteria may exist simultaneously near the center of the wafer. Using the wafer center as the initial search point allows for the acquisition of the maximum scan radius, thereby sampling the matching coordinates of more potential objects. Furthermore, since the stretching deformation direction of the wafer is also from the center outwards, it is possible to sample the matching coordinates of potential objects in regions with different deformation intensities, thus enriching the diversity of the sampled data.

[0065] Therefore, in one embodiment, when multiple candidate potential objects meet the matching conditions, the candidate potential object closest to the center of the wafer can be determined as the starting search point.

[0066] In terms of implementation, the following two methods can be chosen: For example, multiple potential objects within the vicinity of the wafer's center can be identified as candidate potential objects. Then, each candidate potential object can be judged in parallel to determine whether it meets the matching conditions. Finally, the candidate potential object closest to the wafer's center can be selected from the multiple matching objects as the starting search point.

[0067] For example, potential objects can be selected sequentially from the center of the wafer image outwards, in order of increasing distance from the center, and each object can be judged to meet the matching conditions until the first matching object is found and it is determined as the starting search point.

[0068] An exemplary implementation method that uses the starting search point as a seed and identifies potential objects on the corresponding axes based on spatial proximity, progressively identifying search points that satisfy the spatial proximity condition according to the first and second axes respectively: Using the initial search point as a seed, for each iteration, determine the potential objects that exist within the search range of each axis of the current search point, and take the potential object closest to the search point within the search range as the search point for the next round, until the iteration stops when there are no potential objects within the search range of any search point in any round.

[0069] The search points found along each axis are identified as potential objects along the corresponding axis.

[0070] The direction of an axis can be divided into negative and positive directions. For example, the first axis represents the "row direction" of the grain arrangement on the wafer. The first axis can be horizontal to the left with the wafer center as the origin, and horizontal to the right is the first axis positive. The second axis represents the "column direction" of the grain arrangement on the wafer. The second axis can be vertically downward with the wafer center as the origin, and vertically upward is the second axis positive.

[0071] Taking the positive direction of the first axis as an example, for the first... Round-based iterative processing: Determine the initial search point .

[0072] ,by For seeds, determine If a potential object exists within the search range in the positive direction, then the distance within the search range will be [value]. The most recent potential targets are used as search points in the second round. .

[0073] ,Sure If a potential object exists within the search range in the positive direction, then the distance within the search range will be [value]. The most recent potential targets are used as search points in the third round. .

[0074] And so on, Determine the search point If a potential object exists within the search range in the positive direction, then the distance within the search range will be [value]. The most recent potential object is the first Search points of the wheel .

[0075] Determine the search point If a potential object exists within the search range in the positive direction, the iteration process stops.

[0076] like Figure 6 As shown, all potential objects in the positive direction of the first axis can eventually be found (represented by the set of points on the red line). The processing steps for the negative direction of the first axis are similar to those described above and will not be repeated here. Figure 7 As shown, all search points found in the positive and negative directions of the first axis can be identified as potential objects in the first axis (represented by the point set on the red line).

[0077] The process for determining potential objects along the second axis is similar to that for determining potential objects along the first axis, and will not be repeated here.

[0078] In one embodiment, the search range can be a circle with a preset radius centered on the search point. A circular area. Of course, it can also be a rectangular area centered on the search point.

[0079] Optionally, the search range may include a fan-shaped area defined by an angle along a corresponding axial direction with the search point as the center, and the axial direction may include a first axial direction or a second axial direction.

[0080] For example, taking the positive direction of the first axis as an example, the search range can be from the initial search point. and maximum positive and negative deflection angles (Based on process settings, such as 3°, 10°, etc.) Generate two straight lines within a defined range (using Formula 1 and Formula 2). The potential object located within the angle between the two lines and in the positive direction of the first axis... Identify potential objects within the search scope: Formula 1 Formula 2 In this embodiment, since wafer expansion is usually stretched from the center outwards, this stretching not only leads to a larger spacing, but also produces a certain arc or angular displacement. By setting the search range to a fan-shaped area formed by limiting the angle along the corresponding axis with the search point as the center, the path offset caused by the expansion can be covered.

[0081] Periodic dimension refers to the actual physical distance between feature points of two adjacent grains along a specific axis. It essentially represents the sum of the grain's length / width and the kerf width of adjacent grains. Assuming the first axis represents the horizontal direction of the wafer image, the periodic dimension of a grain along this axis can be understood as the sum of the grain's length and the kerf width of adjacent grains. Similarly, assuming the second axis represents the vertical direction of the wafer image, the periodic dimension of a grain along this axis can be understood as the sum of the grain's width and the kerf width of adjacent grains.

[0082] An exemplary implementation of determining the periodic size of a grain in a corresponding axis based on the spacing distribution between the matching coordinates of potential objects in each axis: The periodic size of the grain in the first axis is determined based on the spacing distribution between the matching coordinates of potential objects in the first axis. The periodic size of the grain in the second axis is determined based on the spacing distribution between the matching coordinates of potential objects in the second axis.

[0083] In one embodiment, for any axial direction, based on the spacing distribution between the matching coordinates of adjacent potential objects along the same axial direction, significantly larger spacings, i.e., abnormal spacings, can be identified. The occurrence of abnormal spacing usually indicates that there may be missing grains between the two potential objects corresponding to that spacing (e.g., not identified due to grain defects, blemishes, etc.). For the identified abnormal spacing, a new potential object is inserted between the two potential objects corresponding to the abnormal spacing, and the periodic size of the grain in the corresponding axial direction is determined based on the spacing between the matching coordinates of each adjacent potential object along the interpolated axial direction.

[0084] For example, for the first axis, the anomalous spacing is determined based on the spacing distribution between the matching coordinates of adjacent potential objects along the first axis. A new potential object is inserted between the two potential objects corresponding to the anomalous spacing, and the periodic size of the grain along the first axis is determined based on the spacing between the matching coordinates of adjacent potential objects along the first axis after interpolation.

[0085] For the second axis, the anomalous spacing is determined based on the spacing distribution between the matching coordinates of adjacent potential objects along the second axis. A new potential object is inserted between the two potential objects corresponding to the anomalous spacing. The periodicity of the grain along the second axis is then determined based on the spacing between the matching coordinates of adjacent potential objects after interpolation.

[0086] Taking the first axis as an example, assume that potential objects exist sequentially from the negative direction to the positive direction along the first axis. , , , and The potential object along the first axis after interpolation is , , , , , , .

[0087] When determining the periodic size of the grain in the first axial direction, it is based on , , , , , , The spacing between the matching coordinates is determined.

[0088] In this embodiment, if the spacing between the matching coordinates of adjacent potential objects is too large, it may indicate the presence of missing grains between adjacent potential objects. For example, these grains may not have been identified due to defects, blemishes, or other reasons.

[0089] By detecting abnormal spacing and inserting new potential objects between the two potential objects corresponding to the abnormal spacing, the missing grains in the abnormal spacing can be automatically compensated, so that the grain distribution period after interpolation is complete, thereby improving the reliability of determining the period size of the grains in the corresponding axis.

[0090] Regarding an exemplary implementation for identifying abnormal spacing: For any axis, determine the average distance between adjacent potential objects along the same axis. If the ratio of the distance between any pair of adjacent potential objects to the average distance is not less than a ratio threshold, then the distance between the pair of adjacent potential objects is determined as an abnormal distance.

[0091] For example, for a first axis, the average distance between adjacent potential objects along the first axis is determined. If the ratio of the distance between any pair of adjacent potential objects to the average distance along the corresponding first axis is not less than a ratio threshold, then the distance between the pair of adjacent potential objects is determined to be an abnormal distance.

[0092] For the second axis, determine the average distance between adjacent potential objects along the second axis. If the ratio of the distance between any pair of adjacent potential objects to the average distance along the corresponding second axis is not less than a ratio threshold, then the distance between the pair of adjacent potential objects is determined to be an abnormal distance.

[0093] Taking the first axis as an example, assume that potential objects exist sequentially from the negative direction to the positive direction along the first axis. , , , and .in, and The spacing between them is ; and The spacing between them is ; and The spacing between them is ; and The spacing between them is .

[0094] The average distance between adjacent potential objects can then be determined as: .

[0095] Each spacing can be judged sequentially. Compared with the average proportion .

[0096] If the proportion If the spacing is not less than the proportional threshold, then the spacing will be... The spacing is identified as abnormal. The proportional threshold can be set to 1.5, 1.6, etc.

[0097] An exemplary implementation of inserting a new latent object between two latent objects corresponding to abnormal spacing: Based on proportion The size determines the number of insertions. The size of the ratio is positively correlated with the number of insertions. Insertions are made between two potential objects corresponding to abnormal spacing. A new potential object.

[0098] For example, for any abnormal spacing Number of insertions . Indicates to The result after rounding.

[0099] For example, assuming the ratio threshold is chosen to be 1.5, when the ratio is obtained... It is 1.6, because =1.6>1.5, indicating This is an abnormal spacing; for this abnormal spacing , =2, thus obtaining the number of insertions. =1, meaning in abnormal spacing Insert a new potential object between the two corresponding potential objects.

[0100] Assuming the ratio threshold is chosen to be 1.6, =2.6>1.6, indicating If the spacing is abnormal; then =3, =2, meaning in the abnormal spacing Insert two new potential objects between the two corresponding potential objects.

[0101] In the and the Insert between potential objects The nth potential object, of which the nth The coordinates of the inserted potential objects are: , It should be noted that, .

[0102] In this embodiment, abnormal spacing that is too large can be quickly identified by using the ratio of the spacing to the average spacing. Furthermore, the appropriate number of new potential objects to be inserted can also be determined based on the ratio of the spacing to the average spacing.

[0103] An exemplary implementation of determining the periodic size of a grain in the corresponding axial direction based on the spacing between the matching coordinates of adjacent potential objects in the interpolated axial direction: For example, the average value of the spacing between the matching coordinates of adjacent potential objects in the interpolated axis is determined, and this average value is determined as the periodic size of the grain in the corresponding axis.

[0104] To eliminate potential objects with incorrect spacing, the following improvements were made: Determine the average distance between the matching coordinates of adjacent potential objects along the interpolated axis. If the ratio of the distance between any pair of adjacent potential objects to the average distance is within a preset range, then that distance is determined as the target distance. The average value of the target distance is determined as the periodic dimension of the grain along the corresponding axis.

[0105] For example, determine , , , , , , The average distance between the matching coordinates Determine the spacing between each pair of adjacent potential objects. Compared with the average proportion .like This indicates that the ratio is within the preset range, and the spacing can be adjusted. Determined as the target spacing. If If the ratio is not within the preset range, then the corresponding spacing will not be included in the calculation of the average value. This can exclude abnormal spacing that is too large or too small from being included in the calculation.

[0106] Step 103: Generate the theoretical coordinates of each grain based on the periodic size of the grain and the position index of each grain.

[0107] For example, based on measurement / inspection requirements, the location index of the grains to be generated is read. With the wafer center (0,0) as the origin, calculate the theoretical coordinates of all grains. The grid, such as Figure 8 As shown, The periodic dimension of the grain along the first axis. The periodic dimension of the grain along the second axis.

[0108] Furthermore, this disclosure takes into account that in processes involving physical deformation, such as wafer expansion and dicing, grain distribution exhibits high nonlinear instability. The physical span between grains no longer maintains its initial consistency but fluctuates with changes in spatial position. Taking the wafer expansion process as an example, under the influence of tensile stress, the grain spacing from the center region to the edge region of the wafer will significantly increase, breaking the original equidistant arrangement pattern.

[0109] Therefore, a method is proposed to correct the theoretical coordinates determined based on a unified periodic size, so that the corrected theoretical coordinates can adapt to the nonlinear spatial deformation caused by physical processes, thereby ensuring that the theoretical coordinates are highly consistent with the actual observed coordinates of the grain: In one embodiment, an initial offset between the theoretical coordinates and the matching coordinates of the grains in the central region of the wafer is determined, and all theoretical coordinates are corrected as a whole based on the initial offset.

[0110] The wafer is divided into multiple regions, and each region is processed sequentially from the center outwards.

[0111] Specifically, for the current region to be processed, the grain closest to the wafer center within the current region can be identified as the reference grain, and the local offset between the theoretical coordinates and the matching coordinates of this reference grain can be calculated. Based on the local offset and referring to the arrangement index direction of the reference grain, the theoretical coordinates within the current region are corrected.

[0112] When determining the actual coordinates of each grain based on its theoretical coordinates and the matching coordinates of potential objects in its vicinity, the actual coordinates of each grain can be determined based on its corrected theoretical coordinates and the matching coordinates of potential objects in its vicinity.

[0113] If a grain corresponds to a potential object, that is, the matched potential object is the grain, then the matching coordinates of the grain are the matching coordinates of the potential object.

[0114] When determining the theoretical coordinates of the grains in the central region of a wafer, the theoretical coordinates of the grains closest to the wafer center can be determined.

[0115] In this embodiment, a block-by-block correction method is adopted. Based on the theoretical coordinates and the actual matching coordinates of the neighborhood, the theoretical coordinates of the grains are corrected to prevent large grain offsets at the wafer edge and to effectively remove the cumulative error from the wafer center to the edge caused by grain expansion.

[0116] Regarding the method of dividing the wafer into multiple regions: For example, multiple annular regions can be divided outwards from the wafer center at equal or unequal intervals along the radius. Since the tensile stress caused by film expansion is usually radially symmetrical, the grains within the same annular region have similar degrees of deformation. This method facilitates batch offset compensation for individual grains in each region.

[0117] For example, the wafer can be directly divided into Mesh regions. For certain complex geometric deformations, meshing can better capture local details.

[0118] An exemplary implementation that corrects the theoretical coordinates within the current region based on local offsets and referencing the arrangement index direction of the reference grains: If the arrangement index of the reference grain in the corresponding axis is positive, then based on the local offset, all grains in that axis with arrangement indices greater than the reference grain are corrected.

[0119] If the arrangement index of the reference grain in the corresponding axis is negative, then based on the local offset, all grains in that axis with arrangement indices smaller than the reference grain are corrected.

[0120] In determining the indexing method, the center of the wafer can be set as the origin. Along the first axis, positive indices increase gradually to the right of the wafer center, while negative indices decrease gradually to the left. Along the second axis, positive indices increase gradually upwards from the wafer center, while negative indices decrease gradually downwards.

[0121] For example, for each region, calculate the offset between the theoretical coordinates of the reference grain and the matching coordinates, if the index of that grain... Then for all permutation indices greater than The coordinates of the grain are all involved in the correction; if the index of the grain... Then for all permutation indices less than All coordinates are involved in the correction.

[0122] Similarly, if the index of the grain Then for all permutation indices greater than The coordinates of the grain are all involved in the correction; if the index of the grain... Then for all permutation indices less than All coordinates are involved in the correction.

[0123] For example, with Figure 9Taking the red region 90 as an example, the index of reference grain 91 along the first axis is positive, and the index along the second axis is also positive. Within the red region 90, coordinates with an arrangement index along the first axis greater than the index of reference grain 91 are all included in the correction; within the red region 90, coordinates with an arrangement index along the second axis greater than the index of reference grain 91 are also included in the correction.

[0124] After traversing all regions, the theoretical coordinates have been corrected. Following the above steps, the theoretical coordinates have been corrected, and the theoretical coordinates of the reference grain in each region are now consistent with the matching coordinates.

[0125] In this embodiment, by using the arrangement index direction of the reference grains to correct the theoretical coordinates in the current region, it can be ensured that the compensation correction for the local offset expands from the inside out.

[0126] Step 104: Determine the actual coordinates of each grain based on its theoretical coordinates and the matching coordinates of potential objects in its vicinity.

[0127] In this embodiment, firstly, when generating the theoretical coordinates of the grains, this method does not rely on the nominal periodic size of the grains. Instead, it derives the actual periodic size of the grains based on the spacing distribution between the matching coordinates of potential objects extracted along the mutually perpendicular first and second axes in the wafer image. Then, based on the actual periodic size and the position index of each grain, the theoretical coordinates of the grains are generated, so that the generated theoretical coordinates can reflect the current operating conditions of the wafer. It can be compatible with the distribution of different grains of the same wafer type; it can also be compatible with wafers of different types and processes, and can be applied to uncut wafers and wafer expansion / dicing.

[0128] Secondly, regardless of whether the die is successfully matched, the corresponding actual coordinates can be generated based on the theoretical coordinates of the die and the matching coordinates of potential objects in its vicinity. This makes up for the location of dies that cannot be identified by feature matching. It can be used for wafers with large defects, color differences, or a large number of missing dies, and is compatible with the location of MPW type wafers.

[0129] In one embodiment, for any grain, if there is at least one potential object in the vicinity of the grain's theoretical coordinates, then the matching coordinates of one of the at least one potential object are selected as the actual coordinates of the grain.

[0130] If none exists, the theoretical coordinates of the grain are corrected based on the other theoretical coordinates closest to the theoretical coordinates of the grain and the corresponding matching coordinates, and based on the offset between the other theoretical coordinates and the corresponding matching coordinates, and then used as the actual coordinates of the grain.

[0131] For example, for any grain, such as Figure 10 As shown, the vicinity of this point can be defined as centered on the theoretical coordinates of the grain. For width, A rectangle is generated with height. If there are matching coordinates of at least one potential object within this rectangle, the matching coordinates of one of the potential objects can be selected as the actual coordinates of the grain. For example, if there are multiple potential objects, the matching coordinates of the potential object closest to the theoretical coordinates of the grain can be selected as the actual coordinates of the grain. Of course, if there is only one potential object, the matching coordinates of that potential object can be directly used as the actual coordinates of the grain.

[0132] If it exists, the grain can be marked as a successful match point, the theoretical coordinates of the successful match point can be marked as the theoretical coordinates of the successful match point, and the matching coordinates of potential objects in the vicinity can be used as the matching coordinates corresponding to the theoretical coordinates of the successful match point.

[0133] If no match is found, the grain can be considered a matching failure point, indicating that the actual coordinates of the grain cannot be determined through feature matching. In this case, the theoretical coordinates of the matching failure point can be corrected based on the nearest successful theoretical coordinates and their corresponding matching coordinates, and the offset between the successful theoretical coordinates and their corresponding matching coordinates, and then used as the actual coordinates of the matching failure point.

[0134] For example, from all the successfully matched theoretical coordinates, the theoretical coordinates closest to the failed matching point can be selected, and the theoretical coordinates of the failed matching point can be corrected based on the offset between the closest successful theoretical coordinates and its corresponding matching coordinates, and then used as the actual coordinates of the failed matching point.

[0135] Assume the theoretical coordinates of the matching failure point are The theoretical coordinates of each successfully matched grain that are closest to the point of failure within the theoretical coordinate system. Matching coordinates The actual coordinates of the point where the matching failed can be: , .

[0136] In this embodiment, by using the offset between the theoretical coordinates and the matching coordinates of the successfully matched grains near the matching failure point, the theoretical coordinates of the matching failure point are offset to obtain the actual coordinates of the matching failure point. This ensures that even if the matching fails, the actual coordinates of the matching failure point can still be determined, thereby ensuring comprehensive coverage of subsequent quantity detection.

[0137] Corresponding to the embodiments of the foregoing methods, this specification also provides embodiments of the apparatus and the terminal to which it is applied.

[0138] Figure 11 This is a schematic diagram illustrating the structure of an electronic device according to an exemplary embodiment. Figure 11 As shown, at the hardware level, the electronic device 1100 includes a processor 1102, an internal bus 1104, a network interface 1106, memory 1108, and non-volatile memory 1110, and may also include other hardware required for business operations. One or more embodiments of this specification can be implemented in software, for example, the processor 1102 reads the corresponding computer program from the non-volatile memory 1110 into the memory 1108 and then runs it. Of course, in addition to software implementation, one or more embodiments of this specification do not exclude other implementation methods, such as logic devices or a combination of hardware and software, etc. That is to say, the execution subject of the following processing flow is not limited to each logic module, but can also be hardware or logic devices.

[0139] Figure 12 This is a block diagram illustrating a die positioning device according to an exemplary embodiment of this specification. Figure 12 As shown, this device can be applied to, for example Figure 11 The electronic device 1100 shown implements the technical solution of this specification. The device includes: The matching coordinate determination module 1202 is used to acquire a wafer image and determine the matching coordinates of each potential object in the wafer image through feature matching.

[0140] The period size determination module 1204 is used to determine potential objects located on the first axis and the second axis, and to determine the period size of the grain on the corresponding axis based on the spacing distribution between the matching coordinates of the potential objects on each axis; the first axis and the second axis are perpendicular to each other.

[0141] The theoretical coordinate determination module 1206 is used to generate the theoretical coordinates of each grain based on the period size and the position index of each grain.

[0142] The actual coordinate determination module 1208 is used to determine the actual coordinates of each grain based on the theoretical coordinates of each grain and the matching coordinates of potential objects in its vicinity.

[0143] Optionally, the period size determination module 1204 is used to determine the starting search point; using the starting search point as a seed, on the first axis and the second axis respectively, search points that meet the spatial proximity condition are identified step by step as potential objects on the corresponding axis.

[0144] Optionally, the period size determination module 1204 is used to select potential objects located near the center of the wafer as candidate potential objects; if the candidate potential object meets the matching conditions, the candidate potential object is determined as the starting search point; the matching conditions include: at least one other potential object exists simultaneously within a preset range of the first axis and the second axis of the candidate potential object.

[0145] Optionally, the cycle size determination module 1204 is used to determine the candidate potential object closest to the center of the wafer as the starting search point when multiple candidate potential objects meet the matching conditions.

[0146] Optionally, the period size determination module 1204 is used to determine the potential objects existing in the search range of each axis of the current search point for each iteration, using the starting search point as a seed, and taking the potential object closest to the search point in the search range as the search point for the next round, until there are no potential objects in the search range of any round of search points, and stopping the iteration process; and determining the search points found in each axis as the potential objects in the corresponding axis.

[0147] Optionally, the search range includes a fan-shaped area defined by an angle along a corresponding axial direction with the search point as the center, wherein the axial direction includes a first axial direction or a second axial direction.

[0148] Optionally, the period size determination module 1204 is used to determine the abnormal spacing for any axis based on the spacing distribution between the matching coordinates of adjacent potential objects on the same axis; insert a new potential object between the two potential objects corresponding to the abnormal spacing; and determine the period size of the grain on the corresponding axis based on the spacing between the matching coordinates of each adjacent potential object on the interpolated axis.

[0149] Optionally, the period size determination module 1204 is used to determine the average value of the spacing between adjacent potential objects on the same axis for any axis; if the ratio of the spacing between any pair of adjacent potential objects to the average value is not less than the ratio threshold, then the spacing between the pair of adjacent potential objects is determined as an abnormal spacing.

[0150] Optionally, the period size determination module 1204 is used to determine the number of insertions n based on the ratio of the abnormal spacing to the average value; the size of the ratio is positively correlated with the number of insertions; and n new potential objects are inserted between the two potential objects corresponding to the abnormal spacing.

[0151] Optionally, the period size determination module 1204 is used to determine the average value of the spacing between the matching coordinates of each adjacent potential object in the axial direction after interpolation; if the ratio of the spacing between any pair of adjacent potential objects to the average value is within a preset range, then the spacing is determined as the target spacing; and the average value of the target spacing is determined as the period size of the grain in the corresponding axial direction.

[0152] Optionally, the period size determination module 1204 is further configured to determine the initial offset between the theoretical coordinates and matching coordinates of the grains in the wafer center region, and to perform an overall correction on all theoretical coordinates based on the initial offset; divide the wafer into multiple regions, and process each region sequentially from the center outwards; wherein, for the current region to be processed, determine the reference grain closest to the wafer center in the current region, and calculate the local offset between the theoretical coordinates and matching coordinates of the reference grain; based on the local offset and referring to the arrangement index direction of the reference grain, correct the theoretical coordinates in the current region.

[0153] The theoretical coordinate determination module 1206 is specifically used to determine the actual coordinates of each grain based on the corrected theoretical coordinates of each grain and the matching coordinates of potential objects in its vicinity.

[0154] Optionally, the period size determination module 1204 is further configured to, if the arrangement index of the reference grain in the corresponding axial direction is positive, correct all grains with arrangement indices greater than the reference grain in that axial direction based on the local offset; and if the arrangement index of the reference grain in the corresponding axial direction is negative, correct all grains with arrangement indices less than the reference grain in that axial direction based on the local offset.

[0155] Optionally, the actual coordinate determination module 1208 is configured to, for any grain, if there is at least one potential object in the vicinity of the theoretical coordinates of the grain, select the matching coordinates of one of the at least one potential object as the actual coordinates of the grain; if there is no potential object, then, based on the other theoretical coordinates closest to the theoretical coordinates of the grain and their corresponding matching coordinates, and based on the offset between the other theoretical coordinates and their corresponding matching coordinates, correct the theoretical coordinates of the grain and use them as the actual coordinates of the grain.

[0156] Optionally, the actual coordinate determination module 1208 is used to select the matching coordinates of the potential object closest to the theoretical coordinates from the multiple potential objects as the actual coordinates of the grain if multiple potential objects exist.

[0157] The specific implementation process of the functions and roles of each module in the above device can be found in the implementation process of the corresponding steps in the above method, and will not be repeated here.

[0158] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The modules described as separate components may or may not be physically separate, and the components shown as modules may or may not be physical modules, that is, they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of the solution in this specification according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0159] This specification also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the aforementioned die positioning methods provided in this application.

[0160] Specifically, computer-readable media suitable for storing computer program instructions and data include all forms of non-volatile memory, media, and memory devices, such as semiconductor memory devices (e.g., EPROM, EEPROM, and flash memory devices), magnetic disks (e.g., internal hard disks or removable disks), magneto-optical disks, and CD-ROM and DVD-ROM disks.

[0161] This specification also provides a computer program product, including a computer program / instructions, which, when executed by a processor, implement the steps of any of the aforementioned die positioning methods.

Claims

1. A method of positioning a die, comprising: The method includes: Acquire a wafer image, and determine the matching coordinates of each potential object in the wafer image through feature matching; Potential objects located on the first axis and the second axis are identified, and the periodic size of the grain in the corresponding axis is determined based on the spacing distribution between the matching coordinates of the potential objects on each axis; the first axis and the second axis are perpendicular to each other; The theoretical coordinates of each grain are generated based on the periodic size of the grain and the position index of each grain. The actual coordinates of each grain are determined based on its theoretical coordinates and the matching coordinates of potential objects in its vicinity.

2. The method of claim 1, wherein, The process of identifying potential objects located along the first and second axes includes: Determine the starting search point; Using the initial search point as a seed, search points that satisfy the spatial proximity condition are identified step by step along the first axis and the second axis, respectively, as potential objects on the corresponding axis.

3. The method of claim 2, wherein, Determining the starting search point includes: Potential objects located near the center of the wafer are selected as candidate potential objects; If a candidate potential object meets the matching conditions, the candidate potential object is determined as the starting search point; the matching conditions include: the candidate potential object exists simultaneously with at least one other potential object within a preset range of the first axis and the second axis.

4. The method of claim 3, wherein, If a candidate potential object meets the matching conditions, then the candidate potential object is determined as the starting search point, including: If multiple candidate potential objects meet the matching conditions, the candidate potential object closest to the center of the wafer is determined as the starting search point.

5. The method according to claim 2, characterized in that, The step of using the starting search point as a seed, and progressively identifying search points that satisfy the spatial proximity condition as potential objects on the corresponding axes along the first and second axes, respectively, includes: Using the starting search point as a seed, for each iteration, determine the potential objects that exist within the search range of each axis of the current search point, and take the potential object closest to the search point within the search range as the search point for the next round, until there are no potential objects within the search range of any search point in any round, and stop the iteration process. The search points found along each axis are identified as potential objects along the corresponding axis.

6. The method of claim 5, wherein, The search range includes a fan-shaped area defined by an angle along the corresponding axial direction with the search point as the center, and the axial direction includes a first axial direction or a second axial direction.

7. The method of claim 1, wherein, The determination of the periodic size of the grain in the corresponding axis based on the spacing distribution between the matching coordinates of potential objects in each axis includes: For any axis, the abnormal spacing is determined based on the spacing distribution between the matching coordinates of adjacent potential objects on the same axis; Insert a new potential object between the two potential objects corresponding to the abnormal spacing; Based on the spacing between the matching coordinates of adjacent potential objects along the interpolated axis, the periodic size of the grain along the corresponding axis is determined.

8. The method according to claim 7, characterized in that, For any axial direction, determining the abnormal spacing based on the spacing distribution between the matching coordinates of adjacent potential objects on the same axial direction includes: For any axis, determine the average distance between adjacent potential objects along the same axis; If the ratio of the distance between any pair of adjacent potential objects to the average value is not less than the ratio threshold, then the distance between any pair of adjacent potential objects is determined as an abnormal distance.

9. The method according to claim 8, characterized in that, The insertion of a new potential object between two potential objects corresponding to the abnormal spacing includes: The number of insertions, n, is determined based on the ratio of the abnormal spacing to the average value; the magnitude of this ratio is positively correlated with the number of insertions. Insert n new potential objects between the two potential objects corresponding to the abnormal spacing.

10. The method of claim 7, wherein, The step of determining the periodic size of the grain in the corresponding axial direction based on the spacing between the matching coordinates of adjacent potential objects in the interpolated axial direction includes: Determine the average distance between the matching coordinates of adjacent potential objects along the axis after interpolation; If the ratio of the distance between any pair of adjacent potential objects to the average value is within a preset range, then the distance is determined as the target distance. The average value of the target spacing is determined as the periodic size of the grain in the corresponding axis.

11. The method according to claim 1, characterized in that, The method further includes: Determine the initial offset between the theoretical coordinates and the matching coordinates of the grains in the central region of the wafer, and make an overall correction to all theoretical coordinates based on the initial offset; The wafer is divided into multiple regions, and each region is processed sequentially from the center outwards; Specifically, for the current region to be processed, the reference grain closest to the wafer center within the current region is determined, and the local offset between the theoretical coordinates and the matching coordinates of the reference grain is calculated; based on the local offset and with reference to the arrangement index direction of the reference grain, the theoretical coordinates within the current region are corrected; The process of determining the actual coordinates of each grain based on its theoretical coordinates and the matching coordinates of potential objects within its vicinity includes: The actual coordinates of each grain are determined based on the corrected theoretical coordinates of each grain and the matching coordinates of potential objects in its vicinity.

12. The method of claim 11, wherein, The step of correcting the theoretical coordinates within the current region based on the local offset and with reference to the arrangement index direction of the reference grain includes: If the arrangement index of the reference grain in the corresponding axis is positive, then based on the local offset, all grains in that axis with an arrangement index greater than that reference grain are corrected. If the arrangement index of the reference grain in the corresponding axis is negative, then based on the local offset, all grains in that axis with arrangement indices smaller than the reference grain are corrected.

13. The method of claim 1, wherein, The process of determining the actual coordinates of each grain based on its theoretical coordinates and the matching coordinates of potential objects within its vicinity includes: For any grain, if there is at least one potential object in the vicinity of the theoretical coordinates of the grain, then the matching coordinates of one of the at least one potential object are selected as the actual coordinates of the grain. If none exists, the theoretical coordinates of the grain are corrected based on the other theoretical coordinates closest to the theoretical coordinates of the grain and the corresponding matching coordinates, and based on the offset between the other theoretical coordinates and the corresponding matching coordinates, and then used as the actual coordinates of the grain.

14. The method of claim 13, wherein, The step of selecting the matching coordinates of a potential object from the at least one potential object as the actual coordinates of the grain includes: If multiple potential objects exist, the matching coordinates of the potential object closest to the theoretical coordinates are selected from the multiple potential objects as the actual coordinates of the grain.

15. A device for positioning a die, comprising: The device includes: The matching coordinate determination module is used to acquire a wafer image and determine the matching coordinates of each potential object in the wafer image through feature matching. The period size determination module is used to determine potential objects located on the first axis and the second axis, and to determine the period size of the grain on the corresponding axis based on the spacing distribution between the matching coordinates of the potential objects on each axis; the first axis and the second axis are perpendicular to each other; The theoretical coordinate determination module is used to generate the theoretical coordinates of each grain based on the period size and the position index of each grain; The actual coordinate determination module is used to determine the actual coordinates of each grain based on its theoretical coordinates and the matching coordinates of potential objects in its vicinity.

16. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the method as described in any one of claims 1 to 14.

17. A computer readable storage medium having stored thereon a computer program, characterized in that, When the program is executed by a processor, it implements the steps of the method as described in any one of claims 1 to 14.

Images