A method, apparatus and wafer inspection system for determining wafer scanning strategy

Abstract

This application relates to the field of semiconductor inspection technology, specifically a wafer surface defect detection method, including a wafer scanning strategy determination method, apparatus, and wafer inspection system. The embodiments of this application acquire the distribution results of independent cells on the wafer to be scanned, and adaptively retrieve wafer parameters for subsequent planning based on the distribution status of invalid regions. By constructing the physical spatial coordinate range of independent cells and combining it with key region distribution information to determine the real-time distribution status, the effective regions that need to be scanned can be accurately identified. Furthermore, an image acquisition field-of-view scanning path is generated based on the overlap relationship, ensuring that each independent cell under test is completely covered by the field of view, effectively eliminating the risk of missed detection due to coordinate errors or mechanical drift. Simultaneously, this method employs fully parameterized configuration, which can flexibly adapt to wafers of different sizes and independent cell distributions with different layout densities, enhancing the versatility and adaptability of the technical solution.

Description

Technical Field

[0001] This application relates to the field of semiconductor inspection technology, and is a method for detecting defects on the surface of a wafer, specifically a method, apparatus and system for determining a wafer scanning strategy. Background Technology

[0002] In semiconductor manufacturing processes, wafer inspection is a critical step in ensuring product quality. As chip manufacturing processes continue to shrink, the number of chips integrated on a wafer increases dramatically, placing higher demands on the throughput of inspection equipment. Existing wafer optical inspection technologies typically employ scanning imaging, using a motion platform to move the wafer or lens, in conjunction with an image acquisition device to acquire image data of the wafer surface.

[0003] However, in actual production, due to limitations in upstream process yield and wafer edge effects, the grain distribution on the wafer surface is often uneven, with a large number of invalid or uninspected areas. Existing scanning strategies typically employ full-wafer coverage scanning paths or simply perform rectangular area clipping. When processing wafers with numerous invalid areas, this approach causes the image acquisition device to spend a significant amount of time performing empty scans on these areas, resulting in long idle travel of the motion platform and severely reducing inspection efficiency. Furthermore, existing path planning methods often lack precise calculations of the coverage relationship between grain distribution and the scanning field of view, making them prone to missed detections or duplicate scans due to coordinate errors or mechanical drift, making it difficult to balance inspection efficiency with coverage integrity.

[0004] Therefore, there is an urgent need for a method that can adaptively plan the scan path based on the distribution of independent cells on the wafer, in order to reduce invalid scans and improve detection efficiency. Summary of the Invention

[0005] To address the technical problems in the prior art, embodiments of this application provide a wafer scanning strategy determination method, apparatus, and wafer inspection system. Based on the distribution results of independent units and the real-time distribution status of key areas, the system adaptively generates a field-of-view scanning list and scanning path, achieving skipping of invalid areas and accurate coverage of valid areas, thus significantly improving scanning efficiency.

[0006] To achieve the above objectives, the technical solutions adopted in the embodiments of this application are as follows: In a first aspect, a method for determining a wafer scanning strategy is provided. The method includes: acquiring the distribution results of independent cells on a wafer to be scanned; determining the distribution state of invalid regions on the wafer to be scanned based on the distribution results; and retrieving wafer parameters of the wafer to be scanned when the distribution state does not meet a preset threshold. The wafer parameters include wafer geometric data, independent cell layout data, and image acquisition field-of-view data. Spatial coordinates are constructed for each independent cell to obtain the physical spatial coordinate range of each independent cell. Key region distribution information under standard conditions is acquired; the real-time distribution state of the key regions on the independent cells is determined based on the physical spatial coordinate range of the independent cells and the key region distribution information; the real-time distribution state of the key regions is determined based on the real-time distribution state; and a field-of-view scanning list for the key regions is generated based on the key region distribution state. The key regions are regions on the independent cells. Whether there is an overlap between the spatial coordinate range and the field-of-view scanning list is determined based on the overlap relationship, and an image acquisition field-of-view scanning path covered by the key regions is obtained based on the overlap relationship.

[0007] In some specific implementations, the wafer geometry data includes the wafer radius and the wafer center coordinates in the physical coordinate system; the independent cell layout data includes the independent cell size, independent cell spacing, effective independent cell distribution map, and independent cell start offset; the image acquisition field of view data includes the field of view width, field of view height, field of view overlap rate, and channel parameters; the channel parameters include offset values ​​and redundancy values; and the independent cell layout map is used to characterize the relative positions of the independent cells on the wafer.

[0008] In some specific implementations, spatial coordinates are constructed for each independent unit to obtain the physical spatial coordinate range of each independent unit. This includes: determining a first coordinate and a second coordinate relatively set on the spatial region of each independent unit based on the wafer center coordinates, the wafer radius, the independent unit size, the independent unit spacing, the independent unit starting offset, and the independent unit layout map; determining the center coordinate of the independent unit based on the first coordinate and the second coordinate; and determining the physical spatial coordinate range of the independent space based on the center coordinate and the first and second coordinates.

[0009] In some specific implementations, obtaining key area distribution information under standard conditions and determining the real-time distribution status of the key area on the independent unit based on the physical space coordinate range of the independent unit and the key area distribution information includes: forming a physical space coordinate set by collecting the physical space coordinate range of each independent unit, and traversing the physical space coordinate set and the key area distribution information to determine the independent unit that has an intersection along the second direction as the target independent unit.

[0010] In some specific implementations, determining the real-time distribution status of key areas based on the real-time distribution status includes: pruning the key areas into effective intervals on the target independent unit to obtain a set of effective key intervals; dividing each continuous effective key interval in the set of effective key intervals into multiple region blocks according to the acquisition step size, and determining the overlap relationship between each region block and any independent unit in the physical space coordinate set to obtain the real-time distribution status of the region blocks.

[0011] In some specific implementations, generating a field-of-view scanning sequence for the key region based on the distribution status of the key region includes: obtaining the coverage area of ​​each region block in a first direction, dividing the coverage area based on the field-of-view width to obtain multiple initial field-of-view columns for each region block, sorting them based on a preset scanning path degree and the initial field-of-view columns to obtain a field-of-view list for each region block, and organizing the multiple region blocks to obtain a field-of-view scanning sequence for the key region in a first direction.

[0012] In some specific implementations, obtaining the coverage range of each region block in the first direction includes: determining the center line of each region block in the second direction as a representative row, determining the minimum range in the first direction that needs to be covered on the representative row, and adjusting the minimum range based on the offset value to obtain the coverage range.

[0013] In some specific implementations, determining whether there is an overlap between the spatial coordinate range and the field of view scan list includes: updating the field of view scan list to a two-dimensional matrix to obtain a field of view scan matrix; organizing the physical spatial coordinate range of the independent unit to obtain a two-dimensional matrix about the independent unit; traversing all the field of view scan matrices and the two-dimensional matrix of the independent unit to determine whether there is an overlap between them.

[0014] Secondly, a wafer scanning strategy determination device is provided, the device comprising: a preprocessing module, configured to acquire the distribution results of independent units on a wafer to be scanned, determine the distribution state of invalid regions on the wafer to be scanned based on the distribution results, and retrieve wafer parameters of the wafer to be scanned when the distribution state does not meet a preset threshold; an independent unit processing module, configured to construct spatial coordinates for each independent unit to obtain the physical spatial coordinate range of each independent unit; a field-of-view scanning determination module, configured to acquire key region distribution information under standard conditions, determine the real-time distribution state of the key regions on the independent units based on the physical spatial coordinate range of the independent units and the key region distribution information, determine the real-time distribution state of the key regions based on the real-time distribution state, and generate a field-of-view scanning list for the key regions based on the key region distribution state; and a scanning path determination module, configured to determine whether there is an overlap between the spatial coordinate range and the field-of-view scanning list, and obtain the image acquisition field-of-view scanning path covered by the key regions based on the overlap.

[0015] Thirdly, a wafer inspection system is provided for wafer surface inspection, including an image acquisition device, a memory, and a server; the memory is provided with a configuration file, which includes wafer geometric parameters, scanning field parameters, scanning strategy information, and wafer distribution parameters; the server is used to retrieve the configuration file in the memory and execute the wafer scanning strategy determination method described above to determine the scanning path of the image acquisition device; the image acquisition device acquires a wafer image based on the scanning path and transmits the wafer image to the server.

[0016] The embodiments of the present invention bring the following beneficial effects: The technical solution provided in this application obtains the distribution results of independent cells on the wafer to be scanned, and adaptively retrieves wafer parameters for subsequent planning based on the distribution status of invalid regions, avoiding invalid scanning in invalid regions and significantly improving the efficiency of scanning detection. By constructing the physical spatial coordinate range of independent cells and combining it with the distribution information of key regions to determine the real-time distribution status, the effective regions that need to be scanned can be accurately identified. Furthermore, an image acquisition field-of-view scanning path is generated based on the overlap relationship, ensuring that each independent cell to be tested can be completely covered by the field of view, effectively eliminating the risk of missed detection caused by coordinate errors or mechanical drift. At the same time, this method adopts fully parameterized configuration, which can flexibly adapt to wafers of different sizes and independent cell distributions with different layout densities, enhancing the versatility and adaptability of the technical solution.

[0017] Other features and advantages of this disclosure will be set forth in the following description, or some features and advantages may be inferred from the description or determined without doubt, or may be learned by practicing the techniques described above.

[0018] To make the above-mentioned objects, features and advantages of this disclosure more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] The methods, systems, and / or procedures shown in the accompanying drawings will be further described with reference to exemplary embodiments. These exemplary embodiments will be described in detail with reference to the drawings. These exemplary embodiments are non-limiting exemplary embodiments, wherein example figures represent similar mechanisms in the various views of the drawings.

[0021] Figure 1 This is a schematic diagram of a wafer scanning path determination method.

[0022] Figure 2 This is a schematic diagram showing the distribution results of independent units and the field of view for image acquisition.

[0023] Figure 3 This is a schematic diagram of the distribution results of the regional blocks.

[0024] Figure 4 A schematic diagram of the device structure for determining the wafer scanning strategy.

[0025] Figure 5 This is a schematic diagram of a wafer inspection system.

[0026] Figure 6 This is a schematic diagram of a readable medium structure. Detailed Implementation

[0027] To better understand the above technical solutions, the technical solutions of this application will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments of this application and the specific features in the embodiments are detailed descriptions of the technical solutions of this application, rather than limitations on the technical solutions of this application. In the absence of conflict, the embodiments of this application and the technical features in the embodiments can be combined with each other.

[0028] In the detailed description below, numerous specific details are illustrated with examples to provide a comprehensive understanding of the relevant guidance. However, it will be apparent to those skilled in the art that this application can be practiced without these details. In other instances, well-known methods, procedures, systems, components, and / or circuits have been described at a relatively high level without detail to avoid unnecessarily obscuring aspects of this application.

[0029] This application uses flowcharts to illustrate the execution process performed by a system according to embodiments of this application. It should be clearly understood that the execution processes in the flowcharts may not be executed sequentially. Instead, these execution processes may be executed in reverse order or simultaneously. Additionally, at least one other execution process may be added to the flowchart. One or more execution processes may be deleted from the flowchart.

[0030] This application provides a wafer inspection system for detecting defects on the surface of a wafer, including abnormalities such as breakage, scratches, and contamination. Specifically, the system employs automated optical inspection, acquiring image information of the wafer surface to be inspected and determining whether the wafer has corresponding defects based on pixel features in the image. In this embodiment, "wafer" generally refers to a substrate formed of semiconductor or non-semiconductor materials. Examples include (but are not limited to) single-crystal silicon, gallium arsenide, gallium nitride, and indium phosphide. Such substrates are typically found and / or processed in semiconductor manufacturing facilities. In some cases, a wafer may contain only a substrate (i.e., a bare wafer). Alternatively, a wafer may contain one or more different material layers formed on the substrate. The one or more layers formed on the wafer may be "patterned" or "unpatterned." For example, a wafer may contain multiple bare wafers with repeatable pattern features.

[0031] In this embodiment, the object to be acquired is a single, independent unit. Due to process variations, the configuration of these independent units on the wafer is unique. Therefore, during image acquisition, a corresponding wafer scan path strategy needs to be configured for the current wafer to obtain a complete image.

[0032] Existing wafer scanning strategies mainly fall into two categories: one is to perform pre-planned global path based on the fixed size of the wafer and the position of the calibration center; the other is to first obtain the actual position of the wafer through a positioning camera, and then generate a rectangular scanning area covering the entire wafer based on its outer contour. The essence of both methods is to pursue a "full coverage" scan of the wafer's physical boundaries.

[0033] However, it is worth noting that due to the yield issues of upstream processes such as photolithography and etching, as well as the unavoidable wafer edge effect, there are a large number of invalid or unnecessary areas on the wafer surface. Edge effects can cause defects such as linewidth deviations and pattern residues within a few millimeters of the wafer edge; these areas are typically designated as "invalid regions" and are not used for chip placement. Furthermore, invalid dies may also exist inside the wafer due to process defects. Traditional full-coverage scanning causes the motion platform and image acquisition device to spend a significant amount of time performing "empty scans" on these invalid regions, resulting in long idle travel of the motion platform and severely reducing overall inspection efficiency.

[0034] Furthermore, existing methods lack precise modeling and calculation of the coverage relationship between the independent, discrete effective grain distribution on the wafer and the scanning camera's field of view. Planning is typically based on a simple geometric bounding rectangle, without considering the actual shape, position, and spacing of the grains. This leads to two problems: first, mechanical motion errors, coordinate drift, or misalignment of the field of view may result in some grains not being fully covered by any scanning field of view, causing missed detections; second, to ensure coverage, a large field of view overlap rate is often set, which can easily lead to adjacent fields of view repeatedly scanning the same grain area, increasing data redundancy and processing burden.

[0035] Finally, semiconductor manufacturing involves various processes such as logic chips, memory, and power devices. Different processes have different sensitivities to defects (e.g., logic chips are sensitive to particle defects, while memory is sensitive to pattern offsets), and the surface properties of the wafer materials used (such as silicon, silicon carbide, and gallium nitride) vary. Fixed scan paths and uniform parameters are difficult to adapt to these changes and lack flexibility. When the production line changes products, it often requires downtime to recalibrate the system, increasing non-production time.

[0036] Therefore, in order to improve scanning efficiency and adapt to the process requirements of different product changes, this application provides a wafer scanning strategy determination method. This method aims to solve the problem of low efficiency caused by full scanning of unevenly distributed wafers in the prior art, and achieves accurate detection through adaptive path planning.

[0037] For details on this method, please refer to [link / reference]. Figure 1 Specifically, it includes the following steps: Step S11. Obtain the distribution results of independent cells on the wafer to be scanned, determine the distribution state of invalid regions on the wafer to be scanned based on the distribution results, and retrieve the wafer parameters of the wafer to be scanned when the distribution state does not meet the preset threshold.

[0038] In this embodiment, "independent unit" is a specific expression of the semiconductor technical term "Die." The distribution results of independent units are typically presented as a Die map, characterizing which locations on the wafer surface contain valid chips and which are blank or invalid areas. The distribution of invalid areas refers to the proportion or specific distribution of areas on the wafer surface that do not require detection. A preset threshold is a key indicator for determining the scanning strategy. It should be understood that when the proportion of invalid areas is small, it indicates high wafer surface utilization, and conventional full-wafer coverage scanning will not generate excessive empty passes; in this case, conventional scanning logic can be used. However, when the proportion of invalid areas is large, if full-wafer scanning is still used, the image acquisition device will spend a significant amount of time performing empty scans on the invalid areas, severely reducing detection efficiency. Therefore, this step triggers a subsequent targeted scanning strategy by determining whether the distribution of invalid areas does not meet the preset threshold, demonstrating the flexibility and applicability of the solution. The retrieval of wafer parameters provides the data foundation for subsequent coordinate construction and path calculation.

[0039] Furthermore, in this embodiment, if the distribution ratio of invalid regions is less than 30%, conventional scanning logic is used; if the distribution ratio of invalid regions is greater than 30%, unconventional scanning logic is used. The determination of unconventional scanning logic is the wafer scanning strategy determination method in this embodiment.

[0040] The wafer parameters include circular geometry data, individual cell layout data, and image acquisition field of view data. Specifically, the wafer geometry data includes the wafer radius and the wafer center coordinates in the physical coordinate system. The individual cell layout data includes individual cell size, individual cell spacing, effective individual cell distribution map, and individual cell start offset. The image acquisition field of view data includes field of view width, field of view height, field of view overlap rate, and channel parameters. Channel parameters include offset and redundancy values. The individual cell layout map represents the relative positions of individual cells on the wafer. For details on the configuration relationship between individual cells and the image acquisition field of view, please refer to [reference needed]. Figure 2 As shown.

[0041] Specifically, wafer geometry data defines the physical boundaries of the wafer to be scanned, such as the wafer radius R and center coordinates, which serve as the origin reference for establishing the physical coordinate system. Individual cell layout data describes the arrangement rules of chips on the wafer. The effective individual cell distribution map typically exists in the form of a two-dimensional Boolean matrix, where each element corresponds to an individual cell position on the wafer. Boolean values ​​characterize whether a valid chip exists at that position, thus filtering out the targets to be detected. Image acquisition field of view data defines the single imaging range of the image acquisition device. The field of view width and height determine the basis of the scanning step size, while the field of view overlap rate ensures the continuity of image stitching. In particular, the offset values ​​in the channel parameters are used to compensate for mechanical motion errors or optical system distortions, while the redundancy values ​​ensure that the edges of individual cells are completely covered, preventing missed detections. Both of these factors together improve the robustness of the scanning path.

[0042] Step S12. Construct spatial coordinates for each independent unit to obtain the physical spatial coordinate range of each independent unit.

[0043] In this embodiment, since the initially obtained distribution results are usually logical coordinates (such as row and column indices), they cannot be directly used for the physical positioning of the motion control system. This step maps the logical distribution to the physical coordinate system by introducing wafer geometric data (such as wafer center and radius) and layout data (such as die size and spacing), determining the specific rectangular range of each independent unit in physical space. This process establishes the correspondence between the logical distribution and physical space, providing a prerequisite for subsequent geometric calculations.

[0044] After obtaining the above parameters, the logical distribution needs to be mapped to physical space. Spatial coordinates are constructed for each independent cell to obtain the physical space coordinate range of each independent cell. This includes: determining the first and second coordinates relatively set on the spatial region of each independent cell based on the wafer center coordinates, wafer radius, independent cell size, independent cell spacing, independent cell start offset, and independent cell layout map; determining the center coordinates of the independent cell based on the first and second coordinates; and determining the physical space coordinate range of the independent cell based on the center coordinates and the first and second coordinates.

[0045] Specifically, the coordinate construction process involves converting the logical indices in the independent cell layout map into rectangular regions in the physical coordinate system. For each independent cell (i, j) marked as true in the layout map, its rectangular region range in the physical coordinate system is calculated based on its index (i, j), independent cell sizes Die_size_X and Die_size_Y, and wafer layout rules.

[0046] In this calculation, the first coordinate typically refers to the coordinates of the top-left vertex of the rectangular area of ​​the independent cell, and the second coordinate refers to the coordinates of the bottom-right vertex. The calculation first calculates the relative position based on the index and size interval, then superimposes the wafer center coordinates and the initial offset to obtain the values ​​of the first and second coordinates. For example, the horizontal coordinate of the first coordinate can be calculated using the following logic: wafer center horizontal coordinate + initial horizontal offset + column index × (independent cell width + horizontal interval), and the vertical coordinate is calculated similarly. After obtaining the first and second coordinates, the position and size of the independent cell in physical space are uniquely determined. The center coordinate can then serve as a reference point for subsequent scanning and positioning. It should be understood that this coordinate construction method establishes a precise mapping relationship between the logical distribution and physical space, providing the basic data for geometric calculations to determine the overlap between key areas and the field of view, and avoiding positioning errors caused by inconsistent coordinate systems.

[0047] Step S13. Obtain key area distribution information under standard conditions, determine the real-time distribution status of the key area on the independent unit based on the physical space coordinate range of the independent unit and the key area distribution information, determine the real-time distribution status of the key area based on the real-time distribution status, and generate a field-of-view scan list of the key area based on the key area distribution status.

[0048] In this embodiment, the critical region refers to a specific area on an independent cell that carries the core detection task, such as a specific structural area or vulnerable area inside the chip. The critical region distribution information under standard conditions defines the standard position of the critical region relative to the independent cell. This step combines the standard position with the physical spatial coordinate range of the independent cell to calculate the real-time distribution state of the critical region in the wafer physical coordinate system. Based on this real-time distribution state, the system can generate a field-of-view scan list that only covers the critical region, thereby skipping non-critical regions on the independent cell and invalid regions on the wafer, significantly reducing the number of scan points.

[0049] Specifically, the distribution of the real-time distribution state is formed by combining the physical space coordinate range of each independent unit to form a physical space coordinate set, and by traversing the physical space coordinate set and the key area distribution information, the independent units that have an intersection along the second direction are identified as the target independent units.

[0050] In this embodiment, the key region distribution information under standard conditions typically refers to the standard position definition of the key region relative to the independent cell, such as defining the key region as located in the center or edge region of the independent cell. Since the number of independent cells on a wafer is enormous, performing complex two-dimensional geometric operations directly on each independent cell would consume a large amount of computational resources. Therefore, this embodiment employs an efficient filtering strategy. The system first aggregates the physical space coordinate ranges of all independent cells into a single dataset, and then iterates through this dataset and the key region distribution information. Here, the "second direction" typically refers to the Y-axis direction (vertical direction) of the wafer coordinate system. By determining whether the Y-axis coordinate range of the independent cell intersects with the Y-axis projection of the key region, independent cells that may contain the key region, i.e., target independent cells, can be quickly filtered out. This fast filtering logic along a single direction can eliminate a large number of independent cells that clearly do not contain the key region, thereby significantly reducing the amount of data required for subsequent fine-grained calculations and significantly improving the algorithm's operating efficiency.

[0051] Furthermore, determining the real-time distribution status of key areas based on the real-time distribution status includes: pruning the effective intervals of key areas on the target independent unit to obtain a set of effective key intervals; dividing each continuous effective key interval in the set of effective key intervals according to the acquisition step size to obtain multiple area blocks, and determining the overlap relationship between each area block and any independent unit in the physical space coordinate set to obtain the real-time distribution status of the area blocks.

[0052] This step involves refining the target independent cells. First, the effective range is pruned. Since independent cells at the wafer edge may partially extend beyond the wafer's physical boundary, or the defined range of the critical region may exceed the actual range of the independent cell, directly using the standard-defined critical region would cause the scan path to go out of bounds. Therefore, the system performs an intersection operation between the critical region and the wafer's physical boundary or the independent cell boundary, eliminating the out-of-bounds portions and retaining the actually existing effective range, forming a set of effective critical ranges. This pruning process ensures that the subsequently generated scan path always remains within the wafer's effective physical space, preventing the motion platform from triggering out-of-bounds alarms or invalid movements.

[0053] Secondly, regarding the "segmentation" process, since key regions are typically elongated or irregularly shaped, directly scanning and planning the entire region makes it difficult to control accuracy. The system divides continuous, effective key regions into multiple smaller blocks based on the acquisition step size. The height of each block along the Y-axis is approximately one acquisition step size. This segmentation operation discretizes complex, continuous regions into manageable sub-regions, facilitating independent path planning for each sub-region. Furthermore, the acquisition step size is highly correlated with the image acquisition field of view.

[0054] Finally, the system determines the overlap between each segmented region block and any independent unit in the physical space coordinate set. This step further confirms whether the region block truly covers the independent unit to be detected. If a region block, although located within a critical area, is found not to overlap with any valid independent unit after calculation, it will be marked as invalid or directly discarded. Through this layered screening and fine-grained control, the system ultimately obtains a real-time distribution of region blocks that ensures the integrity of the scan while minimizing invalid scan areas, laying the foundation for generating an efficient field-of-view scan list. For details on the distribution results of the region blocks, please refer to [link to relevant documentation]. Figure 3 As shown.

[0055] Furthermore, for the coverage range of each region block in the first direction, the center line of each region block in the second direction is determined as a representative row, the minimum range in the first direction that needs to be covered on the representative row is determined, and finally the minimum range is adjusted based on the offset value to obtain the coverage range.

[0056] Specifically, the region block is a strip-shaped region divided along the Y-axis, meaning it has a certain height in the second direction. In this embodiment, to simplify computational complexity and avoid tedious coverage calculations for each row of pixels within the region block, the center line of the region block in the Y-axis direction is selected as the representative row. It should be understood that since the height of the region block is usually small, the distribution characteristics of independent units in this representative row on the X-axis can well represent the distribution characteristics of the entire region block. The system traverses all target independent units on the representative row, extracting their minimum and maximum coordinate values ​​in the first direction (X-axis), thereby determining the theoretical minimum coverage area.

[0057] However, in the actual operation of semiconductor testing equipment, the motion platform inevitably experiences mechanical positioning errors, vibrations, or thermal expansion and contraction, and the optical system may also exhibit slight distortions. If scanning is performed only according to the theoretical minimum range, it is highly likely that image gaps will be found in the edge areas of independent units, resulting in missed detections. Therefore, this embodiment introduces an offset value to adjust the minimum range. The offset value is a preset small distance; the system shifts the minimum coordinate value in the opposite direction of the first direction by this value, and shifts the maximum coordinate value in the positive direction of the first direction by this value, thereby expanding the coverage area. This adjustment mechanism effectively compensates for system errors, ensures the robustness of the scanning path, and avoids detection blind spots caused by edge-related missed detections. The specific value of the offset value can be configured according to the accuracy level of the equipment, reflecting the flexibility of the solution.

[0058] Next, a field-of-view scanning sequence for the key region is generated based on the distribution status of the key region, including: obtaining the coverage range of each region block in a first direction, dividing the coverage range based on the field-of-view width to obtain multiple initial field-of-view columns for each region block, sorting them based on a preset scanning path degree and the initial field-of-view columns to obtain a field-of-view list for each region block, and organizing the multiple region blocks to obtain a field-of-view scanning sequence for the key region in a first direction.

[0059] After determining the coverage area, the system needs to discretize it into specific scanning points. The single imaging range of the image acquisition device, i.e., the field of view (FOV), has a fixed width in the first direction, i.e., the X-axis direction. The system divides the adjusted coverage length by the FOV width to calculate how many FOVs are needed to completely cover the region. If the calculation result is not an integer, it is usually rounded up to ensure the integrity of the coverage and avoid uncovered gaps. Thus, each region is divided into several initial FOV columns.

[0060] To further improve detection efficiency and reduce time losses caused by idle travel and frequent starts and stops of the motion platform, the system sorts the initial field-of-view columns according to a preset scanning strategy. Preset scanning strategies include, but are not limited to, unidirectional scanning, zigzag scanning, or serpentine scanning. Taking serpentine scanning as an example, for adjacent regions, if the scanning order of the previous row is from left to right, the scanning order of the next row is automatically adjusted to right to left. In this way, after completing a row of scans, the motion platform does not need to return to the starting point of that row but directly enters the nearest starting point of the next row, significantly reducing the idle travel distance in the X-axis direction and greatly improving the detection throughput. Finally, the system organizes the field-of-view lists of all regions in order to form a complete field-of-view scanning sequence. This sequence contains the precise physical coordinates of each scan point and can be directly sent to the motion control system for execution.

[0061] Step S14. Determine whether there is an overlap between the spatial coordinate range and the field of view scan list, and obtain the image acquisition field of view scan path under the coverage of the key area based on the overlap relationship.

[0062] In this embodiment, determining the overlap relationship is the core step in generating the final scanning path. By judging whether each scanning point in the field-of-view scanning list overlaps with the physical spatial coordinate range of an independent unit, scanning points that fall into invalid or non-critical areas can be eliminated, ensuring that every scanning point in the generated image acquisition field-of-view scanning path is valid. This not only avoids the risk of missed detections but also further optimizes the scanning path, achieving complete avoidance of invalid areas and significantly improving detection efficiency.

[0063] Specifically, to determine the overlap relationship, the field scan list is first updated to a two-dimensional matrix to obtain the field scan matrix. The physical space coordinate range of the independent unit is then organized to obtain a two-dimensional matrix about the independent unit. All field scan matrices and the two-dimensional matrices of the independent units are traversed to determine whether there is an overlap between them.

[0064] In this embodiment, updating the two-dimensional matrix is ​​essentially a process of discretizing and mapping a continuous physical coordinate space. Since the physical coordinates of the wafer surface are continuous, although it is feasible to directly determine the overlap between two continuous geometric regions, the computational load is large and the efficiency is low when there are many independent cells and multiple fields of view.

[0065] Therefore, this embodiment provides a matrix processing logic to solve the above problems. This logic pre-defines a resolution grid, dividing the wafer surface into multiple grid cells. For each field of view in the field-of-view scan list, its physical coordinate range is mapped onto the grid, and the covered grid positions are marked, thereby generating a field-of-view scan matrix composed of 0s and 1s. Similarly, the physical spatial coordinate range of individual cells is also mapped to the same grid coordinate system, generating a two-dimensional matrix of individual cells.

[0066] Furthermore, the above processing method transforms the complex problem of geometric overlap determination into simple matrix logic operations. The system only needs to traverse the two matrices to determine whether the same grid position is marked as valid in both matrices. If so, it indicates that the field of view corresponding to that position overlaps with the independent unit. This matrix-based determination method has significant computational advantages: on the one hand, matrix operations are well-suited for parallel processing in modern computer architectures, significantly reducing computation time; on the other hand, this discretized processing method naturally possesses noise resistance, ignoring minute coordinate fluctuations and improving the system's robustness. It should be understood that although this embodiment uses grid mapping as an example, in practical applications, the matrix can also be a logical matrix based on index relationships, as long as it reflects the spatial correspondence between the field of view and the independent unit. Through this efficient determination algorithm, the system can quickly eliminate those fields of view falling in invalid regions or gaps, retaining only scan points with actual detection significance, thereby ensuring that the final generated image acquisition field of view scanning path is optimized and free of redundancy.

[0067] In summary, this application embodiment obtains the distribution results of independent cells on the wafer to be scanned and adaptively retrieves wafer parameters for subsequent planning based on the distribution status of invalid regions, avoiding invalid scanning in invalid regions and significantly improving the efficiency of scanning detection. By constructing the physical spatial coordinate range of independent cells and combining it with key region distribution information to determine the real-time distribution status, the effective regions that need to be scanned can be accurately identified. Furthermore, the image acquisition field-of-view scanning path is generated based on the overlap relationship, ensuring that each independent cell under test can be completely covered by the field of view, effectively eliminating the risk of missed detection caused by coordinate errors or mechanical drift. At the same time, this method adopts fully parameterized configuration, which can flexibly adapt to wafers of different sizes and independent cell distributions with different layout densities, enhancing the versatility and adaptability of the technical solution.

[0068] For further details, please refer to [link / reference]. Figure 4 This embodiment provides a wafer scanning strategy determination apparatus for executing the aforementioned wafer scanning path determination method. The apparatus employs a modular hardware architecture or software functional unit architecture, enabling efficient processing of path planning tasks during the wafer inspection process.

[0069] Specifically, the wafer scan path determination device 40 includes: The preprocessing module 41 is used to obtain the distribution results of independent cells on the wafer to be scanned, determine the distribution state of invalid regions on the wafer to be scanned based on the distribution results, and retrieve the wafer parameters of the wafer to be scanned when the distribution state does not meet a preset threshold.

[0070] The preprocessing module serves as the data entry point and logical decision-making center of the device. It receives the independent cell distribution results transmitted from the upstream process and analyzes them to identify valid and invalid regions on the wafer surface. This module has built-in threshold comparison logic, capable of calculating the distribution status of invalid regions and comparing it with a preset threshold. When the proportion of invalid regions is determined to be large, the preprocessing module triggers the targeted scanning strategy of this invention, retrieving the corresponding wafer parameters from memory or database, including wafer geometric data, independent cell layout data, and image acquisition field of view data, providing data support for subsequent module calculations.

[0071] Independent unit processing module 42 is used to construct spatial coordinates for each independent unit to obtain the physical spatial coordinate range of each independent unit.

[0072] This module receives wafer parameters from the preprocessing module and performs coordinate mapping. Since the input distribution result is usually a logical index, the independent cell processing module uses a built-in coordinate transformation algorithm, combined with parameters such as wafer center coordinates, independent cell size, and spacing, to map the logical distribution into a physical space coordinate range. This process realizes the transformation from logical layout to physical location, and the output physical space coordinate range is the basic data for subsequent geometric operations.

[0073] The field-of-view scanning determination module 43 is used to acquire key area distribution information under standard conditions, determine the real-time distribution status of the key area on the independent unit based on the physical space coordinate range of the independent unit and the key area distribution information, determine the real-time distribution status of the key area based on the real-time distribution status, and generate a field-of-view scanning list about the key area based on the key area distribution status.

[0074] The field-of-view scanning and determination module is the core computing unit of the device. It first acquires the standard definition of the critical region and, combined with the physical space coordinate range output by the independent unit processing module, calculates the real-time distribution of the critical region on the wafer surface. This module integrates clipping and segmentation algorithms, capable of eliminating invalid regions exceeding the wafer boundary and dividing continuous critical regions into region blocks adapted to the acquisition step size. Subsequently, based on the coverage of the region blocks and the preset scanning strategy, an ordered field-of-view scanning list is generated. This list contains preliminary location information for all points to be scanned.

[0075] The scanning path determination module 44 is used to determine whether there is an overlap between the spatial coordinate range and the field of view scanning list, and to obtain the image acquisition field of view scanning path under the coverage of the key area based on the overlap relationship.

[0076] The scan path determination module, as the output of the device, is responsible for the final optimization of the field-of-view scan list. It checks whether each scan point in the list truly covers a valid independent cell using geometric calculations or matrix-based judgment methods. Redundant scan points falling into invalid areas or gaps are eliminated, ultimately generating an accurate image acquisition field-of-view scan path. This path file can be directly sent to the motion control system to drive the image acquisition device to perform efficient scanning.

[0077] It should be understood that the connections between the above modules represent the flow of data: the output of the preprocessing module is connected to the input of the independent unit processing module, the output of the independent unit processing module is connected to the input of the field of view scanning determination module, and the output of the field of view scanning determination module is connected to the input of the scanning path determination module. In practical applications, these modules can be integrated into servers, industrial control computers, or dedicated image processing cards, and their functions can be implemented through software programs or hardware logic circuits.

[0078] See Figure 5 This embodiment also provides a wafer inspection system 50, which is applied to wafer surface inspection scenarios and aims to achieve high-precision automated scanning and defect detection through the collaborative work of hardware and software. Specifically, the wafer inspection system includes an image acquisition device 51, a memory 52, and a server 5.

[0079] The memory contains a configuration file, which includes wafer geometry parameters, scan field of view parameters, scan strategy information, and wafer distribution parameters. As the system's data foundation, the memory stores configuration files corresponding to various input data required in the aforementioned method embodiments. For example, wafer geometry parameters correspond to geometric data such as wafer radius and center coordinates; scan field of view parameters correspond to field width, height, overlap ratio, and channel parameters; scan strategy information defines the sorting rules for scan paths; and wafer distribution parameters correspond to data such as an independent cell layout map. It should be understood that the memory can be a non-volatile storage medium, such as a hard disk, solid-state drive, or ROM, capable of persistently storing this configuration information even when the device is powered off, ensuring the continuity and reproducibility of the detection task.

[0080] The server retrieves the configuration file from the memory and executes the wafer scanning strategy determination method described in the above embodiments to determine the scanning path of the image acquisition device. The server is the computational core of the system, integrating software algorithm modules such as preprocessing, coordinate construction, field-of-view scanning determination, and path planning. Before the detection task begins, the server reads the configuration file from the memory via the data bus and parses out various parameters. Subsequently, the server determines the distribution status of invalid regions based on the wafer distribution parameters in the configuration file. If the triggering conditions are met, it executes a series of logical operations, including coordinate construction, real-time distribution status calculation of key regions, and overlap relationship determination, ultimately generating an image acquisition field-of-view scanning path containing precise physical coordinates. This scanning path is a set of ordered motion control instructions that guide the movement trajectory of the image acquisition device. It should be noted that the specific form of the server is not limited to a single high-performance computer. Depending on the actual production line layout, it can also be an embedded industrial control computer, a programmable logic controller (PLC), or a cloud computing server, as long as it has sufficient data processing capabilities to execute the aforementioned complex geometric calculation logic.

[0081] The image acquisition device acquires wafer images based on a scanning path and transmits these images to a server. The image acquisition device is the system's execution terminal and typically includes a motion platform (such as an XY-axis precision displacement platform), an optical lens, and an image sensor (such as a CCD or CMOS camera). The scanning path generated by the server is sent to the motion control unit, which drives the motion platform to move the wafer or lens along a planned trajectory. During the movement of the motion platform, the image sensor performs exposure imaging at preset scanning points, acquiring microscopic images of the wafer surface. The acquired image data is transmitted back to the server in real time via a high-speed data transmission interface (such as Camera Link, GigE, or USB 3.0) for subsequent image processing, defect identification, and analysis.

[0082] In terms of the system's interaction sequence, the server first initiates a detection task, reading static configuration data from the memory. After generating dynamic scanning path instructions through internal logic operations, the server sends the instructions to the motion control unit of the image acquisition device. The image acquisition device acquires images in real time during the movement and transmits the data back. This master-slave architecture of "server computing and decision-making, front-end device executing acquisition" decouples computing resources from motion resources, ensuring both the complexity and accuracy of the path planning algorithm and the real-time performance and stability of motion control. Through this system, the technical effects of reducing idle travel and improving detection efficiency in the above-mentioned method embodiments can be perfectly reproduced on a practical hardware platform.

[0083] See Figure 4 The present invention also provides a readable medium 40, which stores computer-readable instructions 401, including instructions for performing the aforementioned wafer scan path determination method.

[0084] The functions and technical effects of the readable medium 40 provided in this embodiment of the invention can be referred to the technical effects in the foregoing embodiments, and will not be repeated here.

[0085] It should be noted that, in the several embodiments provided in this application, it should be understood that the disclosed systems, devices, and / or methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative; for instance, the division of units / modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or modules may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0086] The units / modules described as separate components may or may not be physically separate. The components shown as units / modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units / modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0087] Furthermore, in the various embodiments of the present invention, the functional units / modules can be integrated into one processing unit / module, or each unit / module can exist physically separately, or two or more units / modules can be integrated into one unit / module. The integrated unit / module described above can be implemented in hardware or in the form of hardware plus software functional units / modules.

[0088] The integrated unit / module implemented as a software functional unit / module described above can be stored in a computer-readable storage medium. The software functional unit, stored in a storage medium, includes several instructions to cause one or more processors of a computer device (which may be a personal computer, server, or network device, etc.) to execute some steps of the methods described in the various embodiments of the present invention.

[0089] The integrated unit / module implemented as a software functional unit / module described above can be stored in a computer-readable storage medium. The software functional unit, stored in a storage medium, includes several instructions to cause one or more processors of a computer device (which may be a personal computer, server, or network device, etc.) to execute some steps of the methods described in the various embodiments of the present invention.

[0090] In the description of this invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0091] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for determining a wafer scanning strategy, characterized in that, The method includes: The distribution results of independent cells on the wafer to be scanned are obtained. Based on the distribution results, the distribution state of invalid regions on the wafer to be scanned is determined. When the distribution state does not meet a preset threshold, the wafer parameters of the wafer to be scanned are retrieved. The wafer parameters include wafer geometric data, independent cell layout data, and image acquisition field of view data. Each independent unit is constructed with spatial coordinates to obtain the physical spatial coordinate range of each independent unit; Acquire key region distribution information under standard conditions, determine the real-time distribution status of the key region on the independent unit based on the physical space coordinate range of the independent unit and the key region distribution information, determine the real-time distribution status of the key region based on the real-time distribution status, and generate a field-of-view scan list about the key region based on the key region distribution status; the key region is the region on the independent unit. Based on the spatial coordinate range and the field of view scan list, it is determined whether there is an overlap between the two, and the image acquisition field of view scan path under the coverage of the key area is obtained based on the overlap relationship.

2. The wafer scanning strategy determination method according to claim 1, characterized in that, The wafer geometry data includes the wafer radius and the wafer center coordinates in the physical coordinate system; the independent cell layout data includes the independent cell size, independent cell spacing, effective independent cell distribution map, and independent cell start offset; the image acquisition field of view data includes the field of view width, field of view height, field of view overlap rate, and channel parameters; the channel parameters include offset values ​​and redundancy values, and the independent cell layout map is used to characterize the relative positions of the independent cells on the wafer.

3. The wafer scanning strategy determination method according to claim 2, characterized in that, The spatial coordinates of each independent unit are constructed to obtain the physical spatial coordinate range of each independent unit, including: determining the first coordinate and the second coordinate relative to each independent unit in the spatial region based on the wafer center coordinates, the wafer radius, the independent unit size, the independent unit spacing, the independent unit starting offset, and the independent unit layout map; determining the center coordinate of the independent unit based on the first coordinate and the second coordinate; and determining the physical spatial coordinate range of the independent unit based on the center coordinate and the first coordinate and the second coordinate.

4. The wafer scanning strategy determination method according to claim 2, characterized in that, Obtain key area distribution information under standard conditions, and determine the real-time distribution status of the key area on the independent unit based on the physical space coordinate range of the independent unit and the key area distribution information, including: collecting the physical space coordinate range of each independent unit to form a physical space coordinate set, traversing the physical space coordinate set and the key area distribution information to determine the independent unit that has an intersection along the second direction as the target independent unit.

5. The wafer scanning strategy determination method according to claim 4, characterized in that, Determining the real-time distribution status of key areas based on the real-time distribution status includes: pruning the key areas into effective intervals on the target independent unit to obtain a set of effective key intervals; dividing each continuous effective key interval in the set of effective key intervals into multiple region blocks according to the acquisition step size, and determining the overlap relationship between each region block and any independent unit in the physical space coordinate set to obtain the real-time distribution status of the region blocks.

6. The wafer scanning strategy determination method according to claim 5, characterized in that, Generating a field-of-view scanning sequence for the key region based on the distribution status of the key region includes: obtaining the coverage range of each region block in a first direction, dividing the coverage range based on the field-of-view width to obtain multiple initial field-of-view columns for each region block, sorting them based on a preset scanning path degree and the initial field-of-view columns to obtain a field-of-view list for each region block, and organizing the multiple region blocks to obtain a field-of-view scanning sequence for the key region in a first direction.

7. The wafer scanning strategy determination method according to claim 6, characterized in that, Obtaining the coverage range of each region block in a first direction includes: determining the centerline of each region block in a second direction as a representative row, determining the minimum range in the first direction that needs to be covered on the representative row, and adjusting the minimum range based on the offset value to obtain the coverage range.

8. The wafer scanning strategy determination method according to claim 2, characterized in that, Determining whether there is an overlap between the spatial coordinate range and the field of view scan list includes: updating the field of view scan list to a two-dimensional matrix to obtain a field of view scan matrix; organizing the physical spatial coordinate range of the independent unit to obtain a two-dimensional matrix about the independent unit; traversing all the field of view scan matrices and the two-dimensional matrix of the independent unit to determine whether there is an overlap between them.

9. A wafer scanning strategy determination device, characterized in that, The device includes: The preprocessing module is used to obtain the distribution results of independent cells on the wafer to be scanned, determine the distribution state of invalid regions on the wafer to be scanned based on the distribution results, and retrieve the wafer parameters of the wafer to be scanned when the distribution state does not meet a preset threshold. An independent unit processing module is used to construct spatial coordinates for each independent unit to obtain the physical spatial coordinate range of each independent unit; The field-of-view scanning determination module is used to acquire key area distribution information under standard conditions, determine the real-time distribution status of the key area on the independent unit based on the physical space coordinate range of the independent unit and the key area distribution information, determine the real-time distribution status of the key area based on the real-time distribution status, and generate a field-of-view scanning list about the key area based on the key area distribution status. The scanning path determination module is used to determine whether there is an overlap between the spatial coordinate range and the field of view scanning list, and to obtain the image acquisition field of view scanning path under the coverage of the key area based on the overlap relationship.

10. A wafer inspection system, characterized in that, Applications include wafer surface inspection, such as image acquisition devices, memory, and servers; The memory is configured with a configuration file, which includes wafer geometry parameters, scan field parameters, scan strategy information, and wafer distribution parameters. The server is configured to retrieve the configuration file stored in the memory and execute the wafer scanning strategy determination method according to any one of claims 1-8 to determine the scanning path of the image acquisition device; The image acquisition device acquires a wafer image based on the scanning path and transmits the wafer image to the server.

Images