Design And Implementation Of Leveling System For Lithography Platform Based On Machine Vision

With the rapid development of electronic information technology,the market scale of China's integrated circuit industry is expanding.The demand for chips increases sharply as well.Lithography is the core manufacturing equipment of integrated circuit,its manufacturing precision is the key to guarantee the yield of chips.The leveling system of lithography platform is an important part of lithography,and its leveling precision will directly affect the yield of the chip,the cost and the efficiency of lithography process.At present,the lithography leveling technology usually adopts the refractive and reflective optical path structure,which involves more components,higher overall cost and poor robustness.In this paper,the leveling system of lithography platform is taken as the research object,aiming at solving the robustness problem of the leveling system of lithography platform and improving its adjustment precision,a leveling system of lithography platform based on machine vision is designed.The main research contents are as follows:1)Based on the charge-coupled device(CCD),a unidirectional optical path image acquisition system is designed,and the overall structure design of the leveling system of lithography platform based on CCD imaging is further completed.2)According to the characteristics of high resolution and small field of view of linear CCD images,a preprocessing algorithm based on bilateral filtering,Laplacian sharpening,local adaptive fast binarization and morphological filling is constructed.The extraction and matching of feature points between sampled images and standard images are completed based on ORB--RANSAC algorithm.A screening strategy is designed to extract feature point pairs from matching results that meet the inclination calculation conditions,so as to make them meet the accuracy requirements of inclination calculation,and then achieves the objective of image feature information extraction.3)Aiming at the problem that the tilt angle of platform is difficult to be quantified,a method for calculating the tilt angle based on image information is designed.Firstly,the inclination calculation model of the platform is built based on the principle of vertical projection imaging.Then,the spatial coordinate positions of the feature points are determined through the corresponding relationship between the image feature points matching pairs and the inclination calculation model,and the inclination angle of the platform relative to the horizontal plane could be obtained.At the same time,by calculating the required displacement of the leveling shaft height of the workpiece platform,the leveling scheme of the workpiece platform is designed.4)In this paper,raL8192-12 GM linear CCD camera is used as detection sensor and PCB is used as calibration object.Based on the multi-axis high-precision lithography platform,the leveling system is tested in C++,C# and other software environments.The effectiveness and accuracy of the algorithm module are verified.Compared with the traditional leveling system,the leveling system designed in this paper has the advantages of simple structure and convenient operation,can accurately quantify the inclination value,has more flexible usage scenarios and wider application.