Wafer on-line inspection system

Abstract

The application relates to the technical field of online detection, in particular to a wafer online detection system which comprises an edge calibration module, a probe switching module, a temperature sensing module, an optical detection module and a dynamic compensation module.In the application, the coordinate error of the edge area of the wafer is analyzed in detail, the offset rate data are established, the periodic deviation in the probe switching process is compared, the precision of the probe switching is remarkably improved, the analysis of the temperature sensing and optical reflection data is optimized, and then the accurate spectral layer difference marking diagram is obtained.Through the combination of the regional temperature difference fluctuation spectrum and the optical error diagram, the optical detection and temperature detection of the wafer are dynamically compensated, the wafer online detection effect is greatly improved, the fine positioning and identification of the wafer surface defects are realized, the overall precision and efficiency of the wafer online detection are improved, the risk of defect missed detection and misjudgment is reduced, and the automation level in the production line detection process and the accuracy of the detection result are further optimized.

Description

TECHNICAL FIELD

[0001] The present application relates to the online detection technical field, and particularly to a wafer online detection system. BACKGROUND

[0002] The online detection technical field relates to real-time monitoring and data acquisition of workpiece or material state during product manufacturing or processing, including detection signal acquisition, real-time data analysis, defect identification method, physical quantity conversion and measurement method, and is widely applied to electronic manufacturing, mechanical processing, material forming and other industrial links. The wafer online detection system refers to a system for detecting the wafer surface state under the condition that the wafer does not leave the production line in the semiconductor manufacturing process by using an imaging device and image analysis technology. The wafer surface image during wafer movement is acquired by using a high-speed line array camera, and the image data is compared and analyzed by using image gray distribution feature statistics or wafer pattern matching to identify specific types of surface defects such as scratches, contamination, indentation and particle deposition.

[0003] In the existing wafer online detection process, the surface image is captured by a high-speed camera, and the defect is identified by using image gray distribution or pattern matching method. When facing the change of wafer surface state in a complex environment, it is difficult to cope with the influence of factors such as high real-time requirement error, inaccurate probe switching and temperature fluctuation. Especially, when there is a slight deformation, temperature difference or optical reflection anomaly on the wafer surface, the detection result of the existing technology is prone to error, which reduces the accuracy of defect identification and even causes missed detection, thereby affecting the stability and accuracy of the detection result. SUMMARY

[0004] Therefore, the wafer online detection system is provided to solve the problem that in the prior art, when facing the change of wafer surface state in a complex environment, it is difficult to cope with the influence of factors such as high real-time requirement error, inaccurate probe switching and temperature fluctuation. Especially, when there is a slight deformation, temperature difference or optical reflection anomaly on the wafer surface, the detection result of the existing technology is prone to error, which reduces the accuracy of defect identification and even causes missed detection, thereby affecting the stability and accuracy of the detection result.

[0005] In a first aspect, the wafer online detection system is provided, which comprises:

[0006] The edge calibration module acquires a structure image of the edge region of the wafer to be detected, identifies the optical calibration pattern coordinate group and the temperature sensitive array dot coordinate group in the embedded ring-shaped calibration area boundary, calculates the overall offset value to establish a coordinate error distribution map, compares the error accumulation amount with the total number of points in the region to obtain the offset rate data of the calibration area.

[0007] The probe switching module switches the detection probe to the wafer calibration area based on the offset rate data of the calibration area, records the switching period position change curve, compares the period deviation amount calculated based on the space path length of the calibration area, determines whether the switching forms a backlash behavior, and obtains a probe switching precision index;

[0008] The temperature sensing module activates the resistance array according to the probe switching precision index, collects a sequence of thermal resistance change values of the calibration area, calculates the average thermal response speed of the nodes and performs a response speed difference judgment, marks a temperature sensing uneven behavior if the response speed difference is greater than a response balance limit, and obtains a regional temperature difference fluctuation atlas;

[0009] The optical detection module collects optical reflection images of the marked area based on the marked area in the regional temperature difference fluctuation atlas, obtains a pixel gray scale difference matrix of the current layer and a reference calibration layer at the same coordinate points, locates the position of the maximum absolute value of the pixel difference, and generates a spectral layer difference annotation map.

[0010] The dynamic compensation module performs compensation image reconstruction on the original detection image according to the corresponding relationship between the spectral layer difference annotation map and the regional coordinates in the regional temperature difference fluctuation atlas, and obtains a wafer online detection imaging result map.

[0011] Compared with the prior art, the embodiments of the present application have the beneficial effects that by carefully analyzing the coordinate error of the edge area of the wafer, establishing offset rate data, comparing the period deviation amount in the probe switching process, the precision of the probe switching is significantly improved, and then the analysis of the temperature sensing and optical reflection data is optimized, and then an accurate spectral layer difference annotation map is obtained. Through the combination of the regional temperature difference fluctuation atlas and the optical error map, the optical detection and temperature detection of the wafer are dynamically compensated, the wafer online detection effect is greatly improved, the positioning and recognition of the wafer surface defects are more accurate, the overall precision and efficiency of the wafer online detection are improved, the risk of defect missed detection and misjudgment is reduced, and the automation level and the accuracy of the detection result in the production line detection process are further optimized. BRIEF DESCRIPTION OF DRAWINGS

[0012] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and those skilled in the art can obtain other drawings according to these drawings without creating any creative labor.

[0013] Figure 1 is a system flowchart of the application scenario of the embodiments of the present application;

[0014] Figure 2 is a flowchart of the edge calibration module of the wafer online detection system provided by the embodiments of the present application;

[0015] Figure 3 is a probe switching module flowchart of the wafer online detection system provided by the embodiment of the present application;

[0016] Figure 4 is a temperature sensing module flowchart of the wafer online detection system provided by the embodiment of the present application;

[0017] Figure 5 is an optical detection module flowchart of the wafer online detection system provided by the embodiment of the present application;

[0018] Figure 6 is a dynamic compensation module structure diagram of the wafer online detection system provided by the embodiment of the present application. DETAILED DESCRIPTION

[0019] In the following description, for the purpose of explanation and not limitation, specific details are set forth, such as particular system configurations, techniques, etc., in order to provide a thorough understanding of the embodiments of the present application. However, it will be apparent to those skilled in the art that the present application can be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known systems, devices, circuits, and systems are omitted so as not to obscure the description of the present application with unnecessary detail.

[0020] The wafer online detection system according to the embodiment of the present application will be described in detail below with reference to the accompanying drawings.

[0021] Figure 1 is a flowchart of the wafer online detection system provided by the embodiment of the present application. As shown in the figure, the wafer online detection system comprises: Figure 1

[0022] The edge calibration module acquires a structure image of an edge region of a wafer to be detected, identifies a group of optical calibration pattern coordinates and a group of temperature-sensitive array dot coordinates within a boundary of an embedded ring-shaped calibration area, performs a point-by-point comparison operation on the actual layout template position through coordinate difference accumulation, calculates a pattern overall offset value and establishes an edge region coordinate error distribution map, and obtains calibration area offset rate data by comparing the error accumulation amount in the coordinate error distribution map with the total number of region points.

[0023] The probe switching module drives the piezoelectric ceramic actuator to switch the detection probe to the wafer calibration area based on the calibration area offset rate data, records a position change curve in the switching period and compares the space path length of the calibration area, calculates a period deviation amount using the difference between the position change rate and the set path length, determines whether the switching forms a backlash behavior, and obtains a probe switching precision index.

[0024] ​The temperature sensing module activates the platinum resistance array of the silicon carbide substrate according to the probe switching precision index, collects the thermal resistance change value sequence of each temperature sensing node in the calibration area per unit time, calculates the average thermal response speed of each node, and performs response speed difference value judgment between different regional nodes. If the difference value is greater than the response balance limit, the regional temperature sensing uneven behavior is marked, and the regional temperature difference fluctuation atlas is obtained.

[0025] The optical detection module controls the micro spectrometer probe to enter the marked area to collect the optical reflection image based on the marked area in the regional temperature difference fluctuation atlas, obtains the pixel gray value difference matrix of the current layer and the reference calibration layer at the same coordinate point, performs layer error clustering analysis according to the value range of each pixel point gray value difference, and locates and marks the position corresponding to the maximum point of the pixel value absolute value in each class to generate the spectral layer difference annotation map.

[0026] The dynamic compensation module sets the optical compensation coefficient and the thermal compensation coefficient corresponding to each area as the mapping parameter set according to the corresponding relationship between the spectral layer difference annotation map and the regional coordinates in the regional temperature difference fluctuation atlas, calls the image processing unit in the FPGA module to reconstruct the compensation image of the original detection image zone by zone, and obtains the wafer online detection imaging result graph.

[0027] The high-speed switching mechanism realizes the switching of the detection probe between the wafer working area and the calibration area within 0.1 seconds through the piezoelectric ceramic driving unit. The temperature sensitive array is composed of a platinum resistance film and a silicon carbide substrate, and the thermal response time is ≤5ms. The dynamic compensation module includes an FPGA processor and a compensation parameter storage unit, and configures a temperature-optical double parameter coupling correction algorithm based on the least square method. The detection probe tip is integrated with a micro spectrometer and an infrared temperature sensor, and forms a closed feedback loop with the calibration area.

[0028] The calibration area offset rate data includes optical calibration pattern coordinate group, temperature sensitive array point coordinate group, coordinate difference cumulative value, actual layout template position, pattern overall offset value, edge area coordinate error distribution graph, error cumulative amount, and total point number of the area. The probe switching precision index includes position change curve, calibration area space path length, position change rate, set path length, periodic deviation amount, and back difference behavior record. The regional temperature difference fluctuation atlas includes thermal resistance change value sequence, average thermal response speed, response speed difference value, temperature sensing uneven behavior, and response balance limit. The spectral layer difference annotation map includes optical reflection image, pixel gray value difference matrix, error clustering analysis result, pixel value maximum point, and positioning annotation record. The wafer online detection imaging result graph includes optical compensation coefficient, thermal compensation coefficient, and compensation reconstruction image.

[0029] According to the technical scheme provided in the embodiments of the present application, by analyzing the coordinate error of the wafer edge region in detail, the offset rate data is established, and compared with the periodic deviation in the probe switching process, the precision of the probe switching is significantly improved, and then the analysis of the temperature sensing and optical reflection data is optimized, and then the accurate spectral layer difference marking map is obtained. Through the combination of the regional temperature difference fluctuation map and the optical error map, the optical detection and temperature detection of the wafer are dynamically compensated, the wafer online detection effect is greatly improved, the more fine positioning and recognition of wafer surface defects are realized, the overall precision and efficiency of wafer online detection are improved, the risk of defect missed detection and misjudgment is reduced, and the automation level and the accuracy of the detection result in the production line detection process are further optimized.

[0030] Please refer to Figure 2 , the edge calibration module comprises:

[0031] The edge structure image extraction submodule acquires the edge region structure image of the wafer to be detected, collects the image data of the embedded ring-shaped calibration area as an image input source, performs a boundary recognition operation on the embedded area in the image, detects the pixel positions of the pattern center points and the array point center coordinates by recognizing the pixel distribution characteristics of the embedded optical pattern and the temperature sensitive array points in the ring-shaped calibration area boundary, and performs structure normalization processing on the coordinate positions to generate an initial coordinate set of the embedded pattern;

[0032] Based on the operation of acquiring the edge region structure image of the wafer to be detected by the edge calibration module, first, the imaging device is configured to capture the wafer edge region image to acquire the image data of the embedded ring-shaped calibration area. The ring-shaped calibration area is generally pre-embedded in a region with a radius of 120mm to 125mm at the edge of the wafer, wherein regular distributed optical calibration patterns and temperature sensitive array points are arranged. In actual implementation, an industrial CCD camera with a resolution of 2048x2048 pixels is used for image acquisition. The gray scale range of the obtained image is 0 to 255. After performing normalization processing on the original image data, the brightness interference is eliminated and the contrast is unified. Then, the boundary of the calibration area is recognized in combination with the region characteristics of the image. The region recognition gray scale threshold is set to 190 as the binary segmentation reference. When the pixel gray scale value is greater than the threshold, it is determined as the embedded region boundary pixel, otherwise it is the background. In a certain recognition operation, the edge recognition result of the calibration area shows that the outer edge center position is (x=1024, y=1024) and the radius is 1120 pixels. Based on the boundary coordinate information, the features of the optical patterns and array points in the calibration area are extracted. The center point coordinate set is recognized by judging the symmetry, density and arrangement direction of the pixel set. In one extraction task, 64 optical patterns are detected. The pixel barycenter of each pattern is calculated as the pattern center coordinate. Taking the gray scale distribution of pattern (1, 1) as an example, the x-direction barycenter coordinate is:

[0033] ;

[0034] wherein is the gray value of the pixel in the first row and the first column, and the size of the pattern pixel region is set as , and the gray value of the pixel in the second row and the second column is obtained in the same way, the center pixel coordinates of the pattern are (25.52, 26.40), the operation is repeated to obtain the center coordinates of all patterns and array dots, and then the coordinates are normalized and converted according to the image pixel ratio (1 pixel = 1.5 μm) to obtain the pattern and dot coordinate set in millimeters, in this example, the actual coordinates of the pattern (1, 1) are (0.0383 mm, 0.0396 mm), and finally the initial coordinate set of the embedded pattern is generated.

[0035] Table 1 embedded pattern coordinate extraction result table (unit: pixel and millimeter)

[0036]

[0037] As shown in Table 1, the millimeter coordinate value of the pattern center point can be obtained by normalizing and converting the gray center of gravity coordinates, which provides basic positioning data for subsequent comparison with the template position.

[0038] The calibration coordinate difference generation submodule is based on the initial coordinate set of the embedded pattern and the standard coordinate set of the wafer actually laid template, a pairing relationship is established based on the point index of the two groups of coordinates in the same area, the coordinate difference of the corresponding point is calculated, the coordinate difference of all matched points is accumulated, and the average error coordinate is obtained by dividing the accumulated value by the number of points, a two-dimensional coordinate difference matrix in the edge area of the wafer is established, and a regional coordinate error distribution map is obtained.

[0039] The initial coordinate set of the embedded pattern and the standard coordinate set of the wafer actually laid template are called, first, the positions of the same number of patterns in the two coordinate sets are paired one by one, for example, the position of pattern number (1, 1) in the initial coordinate set is (0.0383 mm, 0.0396 mm), and the preset position in the template standard coordinate set is (0.0400 mm, 0.0400 mm), the offset is obtained by calculating the coordinate difference in the x-axis and y-axis directions of each pair of coordinate points, for pattern (1, 1), the x-axis difference is:

[0040] ;

[0041] Similarly, the y-axis difference is:

[0042] ;

[0043] In the coordinate difference accumulation process, the x-axis and y-axis offset values of all patterns are added point by point. In this example, there are 64 pattern points, and the total difference matrix is set to:

[0044] ;

[0045] The mean value of the matrix is calculated to obtain the overall average error value of the region. Assuming that the total x-axis difference of 64 patterns is -0.102 mm and the total y-axis difference is -0.058 mm, the regional mean error is:

[0046] ;

[0047] ;

[0048] The error is mapped to the position on the wafer edge image, represented by a heat map or a vector arrow indicating the direction and magnitude of the pattern offset, forming a coordinate error vector diagram in two-dimensional space, and establishing a regional coordinate error distribution map.

[0049] The offset rate calculation submodule filters the number of points exceeding the set offset threshold as the error accumulation according to the coordinate difference magnitude of each point in the regional coordinate error distribution map, performs a ratio operation between the error accumulation and the total number of points in the regional coordinate error distribution map, calculates the offset ratio of error points in the calibration region, and obtains the offset rate data of the calibration region;

[0050] According to the coordinate difference magnitude of each point in the regional coordinate error distribution map, the Euclidean distance calculation is performed on the x-axis and y-axis difference of each point to determine whether it exceeds the set offset threshold. The offset distance calculation formula is:

[0051] ;

[0052] Wherein, when mm (set offset threshold) is counted as an offset point. For example, the difference of pattern (1, 1) is (-0.0017, -0.0004), and its offset distance is:

[0053] ;

[0054] Therefore, pattern (1, 1) is not an offset point. Assuming that there are 11 points in 64 points whose offset distance exceeds 0.005 mm, the offset rate calculation formula is:

[0055] ;

[0056] According to the calculation, the offset rate is 17.19%, which means that 17.19% of the pattern points in the calibration area have deviated beyond the standard offset threshold, thus obtaining the calibration area offset rate data.

[0057] Please refer to Figure 3 , the probe switching module comprises:

[0058] The probe target driving sub-module defines the starting point and the ending point of the probe switching path within the boundary range of the wafer calibration area based on the calibration area offset rate data, estimates the error center based on the offset rate data in the area, determines the initial driving direction of the piezoelectric ceramic actuator and the target path curve, guides the piezoelectric ceramic actuator to complete the movement of the probe to the calibration area according to the target path, and records the execution period used for each step in the movement, and generates a probe switching path trajectory data set;

[0059] Based on the calibration area offset rate data, the area with high offset rate value distribution in the wafer surface calibration area is selected as the switching target point set of the piezoelectric ceramic actuator. In a certain detection process, the edge area with an offset rate exceeding 17% is selected as the high offset target area. First, the initial position of the probe is defined as the standby position of the outer circle of the wafer, and the target position is the high offset center point of the calibration area. By analyzing the weight of each coordinate point in the offset rate data matrix, the target path endpoint coordinates are calculated by using the weighted gravity method. In this embodiment, the offset rate weight coefficient is selected as the ratio of the offset rate value of each coordinate point to the total offset rate. It is assumed that there are four key points in the center area of a certain calibration area, and their offset rates are 0.21, 0.18, 0.25 and 0.16. The total offset rate sum is 0.80, and the corresponding coordinates are (2.0, 3.0), (3.0, 3.0), (2.5, 2.5) and (2.0, 2.0). According to the weighted gravity calculation of the target point coordinates, the following is obtained:

[0060] ;

[0061] ;

[0062] The target path endpoint coordinates are (2.4375, 2.7375). Then, based on the target point, the probe switching path is established, the piezoelectric ceramic sheet structure is used as the driving component, the step distance is set to 0.05 mm each time, the control period is 8 ms, the entire path length is 3.5 mm, and the theoretical step period number is 70 times. The system sets the step number marker range to 0 to 69, the control system records the current position coordinates of each step and compares them with the theoretical trajectory curve, constructs a path tracking table, and records the current coordinates, theoretical target points, step numbers, step distances and other information in the table. The following trajectory sampling data is referred to.

[0063] Table 2 Probe switching path trajectory data table

[0064]

[0065] As shown in Table 2, the probe will be slightly offset during the stepping process, and by recording the coordinates of all trajectory points step by step, a complete switching path data structure can be constructed to form a probe switching path trajectory data set.

[0066] The position change acquisition submodule calculates the linear motion rate of the probe in each period using the stepping time and the corresponding position coordinates based on the probe switching path trajectory data set, obtains the speed vector sequence under consecutive periods, compares the difference between the actual moving distance of the probe in each period and the set path length, identifies abnormal pauses and return positions in the path execution process, filters the position change points that have undergone mutations and extracts the corresponding period number, and obtains the probe motion change rate sequence.

[0067] The probe switching path trajectory data set is called to calculate the motion rate of the actual moving trajectory of the probe in each period. First, the actual coordinate difference of any two adjacent points is extracted from the trajectory table, and the average rate is calculated in combination with the stepping period time interval of 8 ms. Taking the first to second periods as an example, the actual positions are (0.049, 0.052) to (0.098, 0.101), and the displacement differences in the X and Y directions are 0.049 mm and 0.049 mm, respectively. The rate calculation is as follows:

[0068] ;

[0069] ;

[0070] The rates of the remaining periods are calculated in the same way to construct a sequence containing the motion rates of consecutive periods. It is assumed that in the consecutive 70 periods, every 8 periods form a monitoring group. By comparing the rate stability in each group, it is determined whether there is a mutation point. If the rate change of any period compared to the previous and subsequent periods exceeds the set reference value of 0.30 mm / s, the period number is recorded as a mutation period. In this embodiment, the rate of the 13th period is 9.2 mm / s, and the rates of the previous and subsequent periods are 8.6 mm / s and 8.4 mm / s, respectively. The fluctuation value of the 13th period is 0.6 mm / s, which is greater than the threshold value, and is considered as an abnormal period. The number is marked as 13, and other mutation periods are continuously counted to form a fluctuation trend table. The numbers and rates of each mutation period are outputted, which are further used for abnormal return difference behavior judgment to obtain the probe motion change rate sequence.

[0071] The period deviation determination sub-module divides the rate difference between any two consecutive periods and performs determination operation according to the probe motion change rate sequence, sets the speed change threshold value as 0.25 mm / s, marks the period number of any period whose rate fluctuation amplitude exceeds the threshold value, calculates the distance difference between the offset path and the target path for these period segments in combination with the path length parameter of the calibration area, judges whether the motion appears a back difference behavior, if the distance difference is greater than the back difference setting threshold value 0.02 mm and the period difference direction is opposite to the path direction, it is regarded as a back difference, the number of back differences is accumulated, and the probe switching precision index is obtained;

[0072] According to the probe motion change rate sequence, difference operation and path difference judgment are performed on all periods marked as rate mutation. First, the actual path coordinate values and the path values of the previous and next periods corresponding to these periods are extracted, and it is judged whether the difference value direction is consistent with the target path direction. If the directions are opposite and the distance difference absolute value is greater than the back difference setting threshold value 0.02 mm, it is recorded as a back difference period. In the 13th period, the coordinates are backtracked from (0.750, 0.751) to (0.700, 0.701), and the path difference is:

[0073] ;

[0074] Since the directions are opposite and the deviation is greater than 0.02 mm, the period is determined as a back difference period. All mutation periods are processed, and 5 back difference periods are identified. In the theoretical path step number 70, the proportion is 7.14%, and the proportion is the switching error rate, which reflects the accuracy of the probe switching process. Further combined with the path execution error distribution graph, the speed sequence and the abnormal point distribution density, a comprehensive index is constructed, and the probe switching precision index of this switching task is obtained.

[0075] Please refer to Figure 4 , the temperature sensing module comprises:

[0076] The thermosensitive node activation sub-module sets a trigger threshold value based on the number of back difference periods and the switching path stability parameter presented in the probe switching process according to the probe switching precision index. When the precision index value exceeds the upper limit of the set range 0.08, the silicon carbide-based platinum resistance array power channel is driven to perform periodic excitation operation on the 128 distributed nodes in the calibration area. The resistance response value sequence is recorded at a period interval of 10 ms and a matrix data structure is established. The resistance change value of each node per unit time is obtained and a response history sequence is constructed, and a thermoresistance change value sequence matrix is generated.

[0077] Based on the probe switching accuracy index, set the driving threshold, when the accuracy index value is greater than 0.08, start the platinum resistance array of silicon carbide substrate, each activation is a single group with 32 nodes, and the total node number is 128, the current excitation period of each node is 10ms after activation, the total measurement period is 1000ms, the system records the resistance change value every 10ms, in a test process, the initial resistance value of the node number N15 is 108.3Ω, the resistance rises to 108.9Ω after 10ms, and reaches 109.5Ω after 20ms, and the resistance value of each node in each time period is obtained by sampling, the sample data structure is constructed as a two-dimensional matrix with time dimension and node dimension as two axes, and the node number is taken as the main index, and the time axis unit is ms step.

[0078] Table 3 thermal resistance change value sequence matrix part sample (unit: Ω)

[0079]

[0080] As shown in Table 3, each node corresponds to a plurality of time resistance values, the sampling resolution is 0.1Ω, each period is collected and stored by the array controller as a data block with a time stamp, the recording process is constant control to the current input, the applied current is 1.0mA, and the sampling stability is guaranteed, the process continues to run until all nodes complete 10 groups of time sampling tasks, after the collection is completed, the original data is classified according to the node number, and each group of resistance data is sorted according to the time index, and a complete thermal resistance change value sequence matrix is formed.

[0081] The thermal response speed calculation sub-module is based on the thermal resistance change value sequence matrix, and the resistance rise amplitude and duration of each node in the continuous sampling period are used for rate calculation. The average resistance increase amplitude in 5 periods is divided by the total time to obtain the response rate value. In a certain node, the resistance rises from 108.2Ω to 109.7Ω, and the duration is 50ms, so the average response rate is 0.03Ω / ms. The 128 nodes are processed in turn to generate a rate array with node index number as the primary key, and the node thermal response speed sequence is obtained.

[0082] The thermal resistance change value sequence matrix is called, the resistance increase amplitude of each node in the set time period is calculated, and the average rate is processed with 5 groups of continuous data as a sliding window. In N15 node data, the resistance increases from 108.3Ω to 110.6Ω from 0ms to 40ms, so the average thermal response rate is:

[0083] ;

[0084] The remaining 127 nodes are processed in the same way to construct a complete thermal response rate array. During each processing, if the node sampling data is less than 5 groups, linear interpolation is used to complete the operation. Subsequently, the rate array is numbered and mapped, each rate value is bound to its corresponding node index number to generate a rate dictionary, and the node ordering sequence is constructed based on the rate dictionary. The maximum and minimum nodes and their rate differences are identified for the next step of regional difference judgment. In this test, the maximum response rate appears at N27, which is 0.063Ω / ms, and the minimum appears at N49, which is 0.025Ω / ms, with a difference of 0.038Ω / ms, which is lower than the threshold value. The current region is not marked as abnormal, and the response rate values of all nodes are written into the storage unit to obtain the node thermal response speed sequence.

[0085] The temperature difference fluctuation determination submodule calculates the rate difference of adjacent regions in the calibration area according to the node thermal response speed sequence. Each four adjacent nodes form a sub-regional unit, and the difference between the maximum rate and the minimum rate in each unit is calculated. If the difference in any sub-region exceeds the set response balance limit of 0.06Ω / ms, the region coordinates are marked and the abnormal type is output. The boundary points of all abnormal regions are counted and superimposed on the wafer area template to obtain the regional temperature difference fluctuation map.

[0086] According to the node thermal response speed sequence, the wafer calibration area is divided into 32 sub-blocks, each containing 4 adjacent nodes. A two-dimensional grid is formed according to the numbering order, and the difference between the maximum rate and the minimum rate of the 4 nodes in each sub-block is calculated as the response imbalance degree index of the sub-region. If the difference is greater than the set response balance limit value of 0.06Ω / ms, the sub-region is marked as an abnormal region. Taking sub-block Z12 as an example, it contains nodes N41-N44, with corresponding thermal response rates of 0.060, 0.065, 0.061 and 0.054Ω / ms, the maximum value is 0.065, the minimum value is 0.054, and the difference is 0.011, which is lower than the set threshold value, and is not marked as abnormal. The node rates in another sub-region Z17 are 0.045, 0.103, 0.050 and 0.048Ω / ms, with a maximum and minimum difference of 0.058, which is slightly lower than the threshold value and does not trigger marking. However, if the actual difference is 0.068, the region Z17 is immediately marked as a temperature difference abnormal region. By superimposing the boundary points of all marked regions on the wafer template map, a thermal response imbalance visualization map is drawn, a two-dimensional color spectrum heat map is constructed, and the regional temperature difference fluctuation map is obtained.

[0087] Referring to Figure 5 , the optical detection module comprises:

[0088] The image acquisition sub-module controls the micro-spectrometer probe to move to the center point coordinates of each region marked in the abnormal region of the regional temperature difference fluctuation atlas and expand the fan-shaped scanning, acquires the reflected light intensity image of each coordinate point under the current layer and synchronously calls the standard image frame of the reference calibration layer, establishes the pixel association matrix in a one-to-one corresponding mapping manner of the original layer and the calibration layer at the spatial coordinate points, and obtains the coordinate mapping image group;

[0089] Based on the abnormal region marked in the regional temperature difference fluctuation atlas, the center point coordinates of all abnormal blocks are extracted by calling the coordinate analysis module according to the coordinate boundary of the red marked region in the atlas, a center point distribution array is constructed, and the center point distribution array is input into the micro-spectrometer probe control unit as a target path. In this process, the probe movement step distance is set to 0.1 mm, the scanning path step range is limited to ±0.2 mm, the probe starts the linear fan-shaped scanning mode at each target region center coordinate point, the scanning angle range is ±30°, and the reflected image of the current region layer is collected within the scanning period. The image acquisition frame rate is set to 20 fps, the image size is 128×128 pixels, after the collection is completed, the reference calibration layer image corresponding to the coordinate point is also loaded synchronously by calling the historical database, a one-to-one mapping table of the pixel points of the current layer and the reference layer at the same spatial coordinate points is established, the mapping table adopts a two-dimensional index structure, each coordinate pair contains the corresponding pixel block number, the gray matrix position identifier and the position matching degree index in the current frame image and the calibration image, and the following is a part of image coordinate mapping data example:

[0090] Table 4: Part of sample data of coordinate mapping image group

[0091]

[0092] As shown in Table 4, a complete spatial mapping relationship is established for each collection point, including the reflected image frame number, the coordinate point position and the matching degree index, and this structure is used for accurate assignment of corresponding pixels in subsequent difference matrix calculation, and finally a coordinate mapping image group is generated.

[0093] The pixel difference matrix calculation submodule extracts the gray values of each pixel point in two layers according to the coordinate mapping image group and performs point-by-point difference calculation, sets the gray value range to 0-255, the current layer gray value at a coordinate point in the abnormal region is 198, the calibration layer gray value is 162, and the corresponding difference value is 36. In this way, the gray difference of all pixel points is calculated, and the formula is:

[0094] ;

[0095] The operation obtains the composite difference amplitude of each pixel point, constructs a two-dimensional matrix structure, and generates a layer pixel gray difference matrix; wherein, represents the first Row The difference value of the column pixel points, is the current layer gray value, is the calibration layer gray value, represents the sum of the transverse gray difference values of the adjacent four points;

[0096] The current image frame and the calibration image frame in the coordinate mapping image group are called, the gray value data of all pixel points in the two-dimensional image matrix is extracted, the gray value interval is limited to 0 to 255, and the gray difference calculation operation is performed on each pair of same-position pixel points. Taking the (i, j) position pixel as an example, if the current image gray value is 198 and the calibration image gray value is 162, the difference value is 36. Further combined with the complexity of the local structure of the image, the gradient change amount of the surrounding pixels is introduced to adjust the difference value. The formula is used for calculation. In a group of actual data, if , , , , , , the calculation process is:

[0097] ;

[0098] The value is the current pixel composite gray difference value. This operation is expanded pixel by pixel on the image matrix, a two-dimensional numerical matrix with the same size as the original image is constructed, a difference value image structure data is formed, and a layer pixel gray difference matrix is obtained.

[0099] The operation logic of the formula can be explained in three parts. First, the first term is the absolute value of the gray difference between the corresponding pixel points of the current image and the reference image , which is used to directly reflect the basic brightness difference between the current layer and the standard layer at the coordinate, and the purpose is to quantify the first-order change trend of the pixel reflection characteristics. The second term is the square root of the product of the gray values of the current layer and the calibration layer. This term is used to express the relative brightness coupling strength between the two gray values. Since the gray values are both non-negative integers, the product is then square-rooted to compress the numerical value to a more suitable numerical scale for superposition with the first term, so as to reflect the influence of the area with high brightness matching degree on the difference value enhancement. The third term is the average value of the gray difference between the current pixel point and its four adjacent pixels in the horizontal direction . This term is used to compensate for the high-frequency fluctuations or noise characteristics existing in the local image. If the current point gray value changes relatively to the adjacent pixel change, the value rises, and the overall difference value forms a deduction term, thereby suppressing the interference of high-frequency errors on the difference image. Therefore, the combination of the three terms is a logic structure of difference value enhancement + matching enhancement - noise cancellation, which can effectively reflect the inter-layer differences of brightness deviation, structure alignment and edge fluctuation.

[0100] The composite difference amplitude refers to the comprehensive difference degree between the current layer and the reference calibration layer in the gray level at the same coordinate pixel point. The difference value not only considers the gray absolute difference value of the corresponding pixel points of the two layers, but also introduces the adjustment factors of brightness coupling strength and local structure disturbance, so that the index can comprehensively reflect the overall state of the brightness deviation of the pixel in the optical imaging, the regional alignment degree and the local texture change. The physical meaning is the overall deviation of a pixel in the structure stability and spectral consistency level relative to the standard image. The larger the value is, the more obvious the optical anomaly or imaging structure disturbance at the position is. The smaller the value is, the stronger the consistency and continuity of the current layer and the calibration layer at the pixel point is.

[0101] The difference label extraction submodule clusters and groups the difference value range according to a fixed hierarchical interval based on the layer pixel gray difference value matrix, divides it into 4 groups according to the gray difference value size, and extracts the coordinate value of the absolute difference value maximum point in each group. After mapping back to the original image space coordinate, red frame labeling is performed and superimposed on the original reflection layer to establish a spectral layer difference label map;

[0102] According to the layer pixel gray difference value matrix, the difference value amplitude in the whole image is range classified and processed, the pixel difference value interval is divided into 4 groups, and the division basis is 0 to 63, 64 to 127, 128 to 191, and 192 to 255. In each section, the maximum value point is selected according to the absolute gray difference value sorting, and its (i, j) position index in the image is recorded. After obtaining the maximum difference point coordinates, the coordinates are backtracked to the corresponding coordinate position of the original image according to the coordinate mapping relationship, and a cross positioning figure is used to perform labeling operation. Each positioning icon is drawn on the upper layer of the image with a red frame. The labeling size is fixed as a 5x5 pixel rectangle, and the position accuracy is ±0.5 pixels. The system generates a set of positioning label coordinate set after each clustering and screening, and performs layer fusion operation on the bitmap rendering layer using bitmap addition mixing method. The marked area in the fused image has high-light display feature, and a spectral layer difference label map is generated.

[0103] Please refer to Figure 6 , the dynamic compensation module comprises:

[0104] The compensation coefficient mapping submodule obtains the coordinate point information in the spectral layer difference label map and the regional temperature difference fluctuation atlas, constructs the mapping relationship structure of the regional coordinates between the two, sets a plurality of optical and thermal compensation coefficient configuration rules according to the joint combination of the gray difference value level distribution and the temperature sensing response rate interval, binds the regional parameter combination results and the regional number index, and generates a regional compensation parameter matrix.

[0105] The coordinate information contained in the spectral layer difference annotation map and the regional temperature difference fluctuation map is obtained, all coordinate points in the two are merged and matched according to the region number, a unified two-dimensional coordinate index table is constructed, and the layer gray difference interval and the temperature difference response rate range corresponding to each region number are extracted, on which the optical-thermal dual parameter combination condition is established, and the combination mapping rule is set: the optical compensation coefficient is set to 0.85 when the gray difference value is between 64 and 127, and the thermal compensation coefficient is set to 1.10 when the temperature difference rate is between 0.05 and 0.07Ω / ms, interval judgment is performed on all region coordinate points and the corresponding compensation coefficient is selected, when there is a region numbered B07, the maximum value of the layer gray difference is 121, and the temperature difference response rate is 0.063Ω / ms, then according to the preset mapping rule, the compensation coefficient combination is selected as 0.85 and 1.10, and recorded as the compensation factor combination of the region, all region compensation factors and region numbers are constructed into a two-dimensional structure matrix, and arranged in ascending order according to the number, and the output result is shown in Table 5:

[0106] Table 5: Example of regional compensation parameter matrix

[0107]

[0108] As shown in Table 5, by putting various physical values in the region into the compensation coefficient mapping relationship, a bidirectional index structure of region number to compensation factor group is established, and a regional compensation parameter matrix is obtained.

[0109] The image dual parameter reconstruction submodule is based on the regional compensation parameter matrix, and according to the gray distribution of each region in the original detection image, the corresponding optical compensation coefficient and thermal compensation coefficient are extracted according to the coordinate index, the correction processing is performed according to the coupling relationship between the two parameters and the original pixel gray, and all pixel points in the region are updated, the compensation result of the whole image is output, and a compensation gray image matrix is obtained;

[0110] The corresponding optical compensation coefficient and thermal compensation coefficient in the regional compensation parameter matrix are called, which are written into the FPGA image processing control buffer according to the region number, and the control unit reads the corresponding region coordinates pixel by pixel before executing the image reconstruction task, and extracts the initial gray value of the pixel in the original detection image, the corresponding coordinate belongs to the region numbered B07, the gray value is 185, and the corresponding compensation coefficient is 0.85 and 1.10, then the system performs subtraction compensation according to the gray value×coefficient weight difference value rule, adjusts the gray value of the current pixel point to obtain the compensated gray value 173, and then processes all pixel points in the region in turn and updates the gray value matrix in real time, if the gray value is modified and exceeds the standard gray upper limit 255 during the matrix updating process, it is set to 255, and if it is lower than 0, it is set to 0, after the processing is completed, the complete image compensation gray matrix is output, and the matrix data stream is packaged in the form of region index structure, and a compensation gray image matrix is obtained.

[0111] The wafer image output sub-module performs a standard gray scale mapping operation on the pixel values according to the compensated gray scale image matrix, uniformly adjusts the pixel values to the range of 0 to 255, performs a truncation reduction process on the part exceeding the range, writes the image cache structure in row and column order and generates a standard 8-bit gray scale image data structure, outputs the whole image data frame and stores it, and obtains the wafer online detection imaging result image;

[0112] According to the compensated gray scale image matrix, a standardization operation of image numerical format is performed, all pixel gray scale values are converted into 8-bit format and expressed in integer form, the upper and lower limit values of the gray scale are set to 0 and 255 in the standardization process, and the pixels exceeding the range value are subjected to a truncation process, then the whole image is re-encoded into an image buffer frame format in row and column order and written into the image display output end buffer channel in the FPGA, a buffer pointer refresh is performed in the buffer area and the current compensated image frame is output, the image data is converted into a standard output image format PNG by the video stream protocol conversion module and written into the image frame recording unit, the image structure contains complete area number index, position coordinate mapping structure and gray scale value matrix, a unified structure data file output result is finally generated, and the wafer online detection imaging result image is obtained.

[0113] All the optional technical solutions described above can be combined to form optional embodiments of the present application, which will not be described one by one here.

[0114] The above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements to some technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. A wafer online inspection system, characterized in that, The system includes: The edge calibration module acquires the structural image of the edge region of the wafer to be inspected, identifies the coordinate group of the optical calibration pattern and the coordinate group of the temperature-sensitive array within the boundary of the embedded ring calibration area, calculates the overall offset value to establish a coordinate error distribution map, and compares the cumulative error with the total number of points in the region to obtain the offset rate data of the calibration area. Based on the calibration area offset rate data, the probe switching module switches the detection probe to the wafer calibration area, records the switching cycle position change curve, compares the cycle deviation with the spatial path length of the calibration area, determines whether the switching has formed hysteresis behavior, and obtains the probe switching accuracy index. The temperature sensing module activates the resistor array according to the probe switching accuracy index, collects the thermal resistance change value sequence of the calibration area, calculates the average thermal response speed of the node and performs response speed difference judgment. If it is greater than the response equilibrium limit, it is marked as temperature sensing uneven behavior, and the regional temperature difference fluctuation map is obtained. The optical detection module acquires the optical reflection image of the marked area based on the marked area in the temperature difference fluctuation spectrum of the region, obtains the pixel gray level difference matrix between the current layer and the reference calibration layer at the same coordinate point, locates the position of the point with the maximum absolute value of the pixel difference, and generates a spectral layer difference annotation map. The dynamic compensation module performs compensation image reconstruction on the original detection image region by region based on the correspondence between the spectral layer difference annotation map and the regional coordinates in the regional temperature difference fluctuation map, and obtains the wafer online detection imaging result map.

2. The wafer online inspection system according to claim 1, characterized in that, The calibration area offset rate data includes the optical calibration pattern coordinate group, the temperature-sensitive array point coordinate group, the cumulative coordinate difference value, the actual template position, the overall pattern offset value, the edge area coordinate error distribution map, the cumulative error amount, and the total number of points in the area; the probe switching accuracy index includes the position change curve, the calibration area spatial path length, the position change rate, the set path length, the period deviation amount, and the hysteresis behavior record; the area temperature difference fluctuation spectrum includes the thermal resistance change value sequence, the average thermal response rate, the response rate difference, the temperature sensing unevenness behavior, and the response balance limit. The spectral layer difference annotation map includes optical reflection image, pixel gray-level difference matrix, error clustering analysis results, pixel difference maximum point, and location annotation record; The wafer online inspection imaging results include optical compensation coefficient, thermal compensation coefficient, and compensated reconstructed image.

3. The wafer online inspection system according to claim 1, characterized in that, The edge calibration module includes: The edge structure image extraction submodule acquires the edge region structure image of the wafer to be detected, collects embedded ring calibration area image data as image input source, performs boundary recognition operation on the embedded region in the image, and detects the pixel position of the pattern center point and the center coordinate of the temperature sensitive array by identifying the pixel distribution characteristics of the embedded optical pattern and temperature sensitive array points within the boundary of the ring calibration area, and performs structural normalization processing of the coordinate position to generate the initial coordinate set of the embedded pattern. The calibration coordinate difference generation submodule establishes a pairing relationship based on the initial coordinate set of the embedded pattern and the standard coordinate set of the actual wafer layout template, based on the point index of the two sets of coordinates in the same area. By calculating the coordinate difference of the corresponding points, the coordinate difference accumulation operation is performed on all matching points, and the accumulated value is divided by the number of points to obtain the mean error coordinate. A two-dimensional coordinate difference matrix is ​​established in the wafer edge area to obtain the regional coordinate error distribution map. The offset rate calculation submodule selects the number of points that exceed the set offset threshold as the cumulative error based on the coordinate difference of each point in the regional coordinate error distribution map. It then calculates the offset ratio of the error points in the calibration area by performing a ratio calculation with the total number of points in the regional coordinate error distribution map, and obtains the offset rate data of the calibration area.

4. The wafer online inspection system according to claim 1, characterized in that, The probe switching module includes: Based on the calibration area offset rate data, the probe target driving submodule defines the starting point and ending point of the probe switching path within the boundary of the wafer calibration area. It estimates the error center through the offset rate data within the area, determines the initial driving direction and target path curve of the piezoelectric ceramic actuator, guides the piezoelectric ceramic actuator to complete the probe transfer action to the calibration area according to the target path, and records the execution cycle used for each step in the transfer, generating a probe switching path trajectory data set. The position change acquisition submodule calculates the linear motion rate of the probe in each cycle using the step time and corresponding position coordinates based on the probe switching path trajectory data group, obtains the velocity vector sequence in continuous cycles, compares the difference between the actual moving distance of the probe in each cycle and the set path length, identifies abnormal pauses and return positions during path execution, filters out position change points that have abrupt changes and extracts the corresponding cycle number to obtain the probe motion change rate sequence. The period deviation determination submodule performs interval division and determination calculation on the rate difference between any consecutive periods based on the probe motion change rate sequence, marks the period number where the rate fluctuation amplitude exceeds the speed change threshold, and calculates the distance difference between the offset path and the target path for the corresponding period segment in combination with the calibration area path length parameter, determines whether there is backlash behavior during the movement, accumulates all backlash counts, and obtains the probe switching accuracy index.

5. The wafer online inspection system according to claim 4, characterized in that, The process of marking the period number in which the rate fluctuation amplitude of any period exceeds the threshold is as follows: the absolute value of the rate difference between two consecutive periods is compared with the speed change threshold. If the absolute value of the rate difference is greater than or equal to the speed change threshold, the corresponding period is marked as an abnormal period.

6. The wafer online inspection system according to claim 5, characterized in that, The process of determining whether there is backlash behavior during movement is as follows: In the abnormal cycle, if the angle between the movement direction of the current cycle and the direction of the target path curve is greater than 90 degrees, and the distance difference between the offset path and the target path is greater than the backlash threshold, then it is determined that a backlash has occurred.

7. The wafer online inspection system according to claim 1, characterized in that, The temperature sensing module includes: Based on the probe switching accuracy index, the thermal node activation submodule drives the power channel of the silicon carbide substrate platinum resistance array when the accuracy index value exceeds the trigger threshold, according to the hysteresis cycle number and switching path stability parameters presented during the probe switching process. It performs periodic excitation operations on the distributed nodes in the calibration area, records the resistance response value sequence and establishes a matrix data structure, obtains the resistance change value of each node per unit time and constructs the response history sequence, and generates a thermal resistance change value sequence matrix. The thermal response rate calculation submodule calculates the rate based on the thermal resistance change value sequence matrix and the resistance increase amplitude and duration of each node in the continuous sampling period. The response rate value is obtained by dividing the average resistance increase by the total duration. All nodes are processed in sequence to obtain a rate array with the node index number as the primary key, and the node thermal response rate sequence is obtained. The temperature fluctuation determination submodule calculates the rate difference between adjacent regions within the calibration area based on the node thermal response rate sequence. Every four adjacent nodes constitute a sub-region unit. The difference between the maximum and minimum rates within each unit is calculated. If the difference within any sub-region exceeds the set response equilibrium limit, the coordinates of that region are marked and the anomaly type is output. The boundary points of all anomaly regions are counted and superimposed onto the wafer region template to obtain the regional temperature fluctuation map.

8. The wafer online inspection system according to claim 1, characterized in that, The optical detection module includes: Based on the abnormal areas marked in the temperature difference fluctuation spectrum of the region, the image acquisition submodule controls the micro spectrometer probe to move to the center point coordinates of each region and unfolds a fan-shaped scan, acquires the reflected light intensity image of each coordinate point in the current layer and synchronously calls the standard image frame of the reference calibration layer, and establishes a pixel correlation matrix by using a one-to-one mapping method between the original layer and the calibration layer in spatial coordinate points to obtain the coordinate mapping image group. The pixel difference matrix calculation submodule extracts the gray value of each pixel in the two layers according to the coordinate mapping image group and performs point-by-point difference calculation. It sets the gray value range to calibrate the gray value of the layer, calculates the gray value difference of all pixels, calculates the composite difference amplitude of each pixel, constructs a two-dimensional matrix structure, and generates the layer pixel gray value difference matrix. The difference annotation extraction submodule performs clustering and grouping of the difference value range based on the pixel gray-level difference matrix of the layer using a fixed hierarchical interval. It divides the group according to the size of the gray-level difference and extracts the coordinate value of the point with the largest absolute difference in each group. After mapping back to the original image space coordinates, it marks the difference with a red box and superimposes it onto the original reflection layer to establish a spectral layer difference annotation map.

9. The wafer online inspection system according to claim 1, characterized in that, The dynamic compensation module includes: The compensation coefficient mapping submodule obtains the coordinate point information in the spectral layer difference annotation map and the regional temperature difference fluctuation map, constructs the mapping relationship structure between the two regional coordinates, sets multiple sets of optical and thermal compensation coefficient configuration rules according to the joint combination of gray level difference distribution and temperature response rate interval, binds the parameter combination results of each region with the region number index, and generates a regional compensation parameter matrix. The image dual-parameter reconstruction submodule, based on the region compensation parameter matrix, extracts the corresponding optical compensation coefficient and thermal compensation coefficient according to the coordinate index of each region in the original detection image, performs correction processing according to the coupling relationship between the two parameters and the original pixel gray level, updates all pixels in the region, outputs the whole image compensation result, and obtains the compensated gray level image matrix. The wafer image output submodule performs a standard grayscale mapping operation on the pixel values ​​according to the compensated grayscale image matrix, uniformly adjusting the pixel values ​​to the range of 0 to 255. The parts exceeding the range are truncated and reduced. The data is written into the image cache structure in row and column order and a standard grayscale image data structure is generated. The entire image data frame is output and stored to obtain the wafer online inspection imaging result image.

Images