CN-122027894-A - Automatic focusing method, focusing system, computer readable storage medium and electronic device for wafer detection

CN122027894ACN 122027894 ACN122027894 ACN 122027894ACN-122027894-A

Abstract

The invention provides an automatic focusing method, a focusing system, a computer readable storage medium and electronic equipment for wafer detection, wherein the automatic focusing method comprises the steps of collecting original images with different exposure times through a camera for a wafer surface focusing area, performing high dynamic range HDR synthesis and nonlinear correction to obtain a preprocessed image, calculating at least one quality evaluation index of the preprocessed image, judging that a current image is invalid and abandoned if the index does not reach a preset threshold value, executing subsequent steps if the index passes verification, extracting at least one focusing definition characteristic from the preprocessed image passing verification, calculating to obtain a comprehensive focusing score of the current image based on the extracted focusing definition characteristic, searching in the Z-axis direction, repeatedly executing the steps at different Z-axis positions to obtain the comprehensive focusing score corresponding to each position, and searching for a Z-axis position which enables the comprehensive focusing score to reach the maximum as an optimal focusing position.

Inventors

SHI ZHUBIN
XU ZHIDA
LIU XIANGHUA

Assignees

麦峤里(上海)半导体科技有限责任公司

Dates

Publication Date: 20260512
Application Date: 20260414

Claims (10)

1. An auto-focus method for wafer inspection, comprising: S1, acquiring original images with different exposure times for a focusing area on the surface of a wafer through a camera, and performing high dynamic range HDR synthesis and nonlinear correction to obtain a preprocessed image; S2, calculating at least one quality evaluation index of the preprocessed image, if the index does not reach a preset threshold, judging that the current image is invalid and is discarded; s3, extracting at least one focusing definition characteristic from the preprocessed image passing verification; step S4, calculating and obtaining the comprehensive focusing score of the current image based on the extracted focusing definition characteristics; And S5, searching in the Z-axis direction, repeatedly executing the steps S1 to S4 at different Z-axis positions iteratively to obtain comprehensive focusing scores corresponding to the positions, and searching and positioning the Z-axis position which enables the comprehensive focusing score to reach the maximum by comparing the scores of the positions to be used as the optimal focusing position.
2. The auto-focus method for wafer inspection according to claim 1, wherein in step S1, the performing high dynamic range HDR synthesis and nonlinear correction comprises: reconstructing a linear irradiance map of the scene using the original images of the plurality of frames of different exposure times based on Debevec algorithm; And performing nonlinear mapping based on gamma correction on the linear irradiance map, and quantizing a mapping result into an image with 8 bit depth to serve as the preprocessing image.
3. The method according to claim 1, wherein in step S2, the quality evaluation index includes average brightness, brightness dispersion and overexposure saturation; If the average brightness is smaller than a first threshold value or larger than a second threshold value, judging that the exposure is abnormal and invalid; if the brightness dispersion is smaller than a third threshold value, judging that the image is an invalid image with low contrast; And if the overexposure saturation rate is greater than a fourth threshold value, judging that the image is an overexposed invalid image.
4. The auto-focus method for wafer inspection according to claim 1, wherein in step S3, the focus sharpness features include high frequency energy features, edge density features, and texture entropy features; The high-frequency energy characteristic is obtained by calculating the variance of the response value of the preprocessed image after Laplace convolution operation is carried out on the preprocessed image; The edge density characteristic is obtained by carrying out Canny edge detection on the preprocessed image and calculating the proportion of edge pixels to total pixels; And the texture entropy features are obtained by constructing a gray level co-occurrence matrix of the preprocessed image, normalizing the co-occurrence matrix to obtain a probability matrix and calculating the entropy of the probability matrix.
5. The auto-focusing method for wafer inspection according to claim 4, wherein in step S3, the high-frequency energy features, the edge density features, and the texture entropy features are extracted only from a central region of interest of the pre-processed image; the central interested region is a region with the geometric center of the preprocessed image as the center, and the width and the height are respectively a certain proportion of the width and the height of the original image.
6. The method for auto-focusing for wafer inspection according to claim 4, wherein step S4 comprises: Respectively carrying out normalization processing on the high-frequency energy characteristic, the edge density characteristic and the texture entropy characteristic to obtain normalized characteristic values; And carrying out weighted fusion on the normalized characteristic values according to the process layer type of the wafer, and calculating the comprehensive focusing score.
7. The method for auto-focusing for wafer inspection according to claim 1, wherein in step S5, fibonacci searching is adopted for the searching in the Z-axis direction; The fibonacci searching method searches a position which maximizes the comprehensive focusing score through iterative comparison and interval contraction, and terminates searching when the length of a section is smaller than or equal to a set threshold value or the lifting of the comprehensive focusing score is smaller than the set threshold value or the total sampling point number reaches a preset value, and outputs the optimal focusing position and the corresponding maximum comprehensive focusing score.
8. An autofocus system for wafer inspection, comprising: The image acquisition module is used for acquiring original images of a plurality of frames with different exposure time; The image preprocessing module is used for carrying out high dynamic range HDR synthesis and nonlinear correction on the acquired original image to obtain a preprocessed image; The image scoring module is used for calculating at least one quality evaluation index of the preprocessed image, extracting at least one focusing definition characteristic from the preprocessed image which passes verification, and calculating and obtaining the comprehensive focusing score of the current image based on the extracted focusing definition characteristic; And the focusing module is used for controlling the Z-axis motion platform to obtain comprehensive focusing scores corresponding to all the positions, and searching and positioning the Z-axis position which enables the comprehensive focusing score to reach the maximum by comparing the scores of all the positions to be used as the optimal focusing position.
9. A computer readable storage medium, on which a computer program is stored, which computer program, when being executed by a processor, implements the method according to any one of claims 1 to 7.
10. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the method of any one of claims 1 to 7 when the program is executed by the processor.

Description

Automatic focusing method, focusing system, computer readable storage medium and electronic device for wafer detection Technical Field The present invention relates to the field of semiconductor manufacturing and inspection technology, and in particular, to an auto-focusing method, a focusing system, a computer readable storage medium, and an electronic device for optical inspection of a wafer. Background In semiconductor manufacturing processes, the surface quality of wafers (e.g., photolithographic patterns, film thicknesses, defects, etc.) needs to be monitored by high resolution vision inspection systems. The accuracy of the detection is directly dependent on the sharpness of the acquired image. In actual production, the thickness of the wafer often has fine deviation, so that the detection camera cannot be continuously measured at the same height all the time in the measurement process, the height needs to be changed in real time, the imaging effect of the detection camera is required to be detected, the imaging is clear, and the follow-up measurement is convenient. In order to obtain a clear image, an auto-focus technology becomes one of core functions of the high-end wafer inspection apparatus. The conventional automatic focusing method is generally based on a definition evaluation function of an image, such as a gradient function, a frequency domain function, and the like, and finally searches for the position with the highest score as the best focusing point by driving a Z-axis motion platform to scan different heights and calculating the definition score of each image. During automatic focusing, the metal layer on the surface of the wafer has strong specular reflection characteristics, so that a single frame image is extremely easy to be partially overexposed or underexposed, a traditional definition evaluation algorithm is disabled, image quality (such as overall darkness and low contrast) fluctuation caused by illumination and process layer difference can interfere with the reliability of a focusing criterion, and besides, aberration (field curvature and distortion) at the edge of an optical lens can introduce irrelevant blurring to influence the accuracy of the definition scoring of the whole image. Therefore, how to find the highest scoring position as the best focus is a problem that needs to be solved at present. Disclosure of Invention The invention aims to provide an automatic focusing method for wafer detection, which can find the optimal focusing position. In order to achieve the above object, the present invention provides an auto-focusing method for wafer inspection, comprising: S1, acquiring original images with different exposure times for a focusing area on the surface of a wafer through a camera, and performing high dynamic range HDR synthesis and nonlinear correction to obtain a preprocessed image; S2, calculating at least one quality evaluation index of the preprocessed image, if the index does not reach a preset threshold, judging that the current image is invalid and is discarded; s3, extracting at least one focusing definition characteristic from the preprocessed image passing verification; step S4, calculating and obtaining the comprehensive focusing score of the current image based on the extracted focusing definition characteristics; And S5, searching in the Z-axis direction, repeatedly executing the steps S1 to S4 at different Z-axis positions iteratively to obtain comprehensive focusing scores corresponding to the positions, and searching and positioning the Z-axis position which enables the comprehensive focusing score to reach the maximum by comparing the scores of the positions to be used as the optimal focusing position. In an alternative, in step S1, the performing high dynamic range HDR synthesis and nonlinear correction includes: reconstructing a linear irradiance map of the scene using the original images of the plurality of frames of different exposure times based on Debevec algorithm; And performing nonlinear mapping based on gamma correction on the linear irradiance map, and quantizing a mapping result into an image with 8 bit depth to serve as the preprocessing image. In the optional solution, in step S2, the quality evaluation index includes average brightness, brightness dispersion and overexposure saturation rate; If the average brightness is smaller than a first threshold value or larger than a second threshold value, judging that the exposure is abnormal and invalid; if the brightness dispersion is smaller than a third threshold value, judging that the image is an invalid image with low contrast; And if the overexposure saturation rate is greater than a fourth threshold value, judging that the image is an overexposed invalid image. In an alternative scheme, in step S3, the focusing definition features include a high-frequency energy feature, an edge density feature and a texture entropy feature; The high-frequency energy characteristic is obtained by calc