Inspection And Evaluation Of Lens Quality For High-end Packaging Equipment Based On Modulation Transfer Function

With the development and progress of microelectronic packaging technology,integrated circuit packaging technology is developing towards high density and high complexity,and the packaging performance requirements for chip packaging are constantly improving.Among them,the quality of optical lenses is one of the performance indicators that determine the performance of semiconductor packaging equipment.Modulation Transfer Function(MTF)is an important indicator widely used to qualitatively and quantitatively express the performance of an optical imaging system.The MTF curve displays both resolution and contrast information,enables the evaluation of lenses based on the needs of a specific application,and can be used to compare the performance of multiple lenses.This thesis researches on the quality inspection of the lens module packaged in the production line.Through theoretical analysis and practice,comparing various calculation methods,the improved algorithm is proposed to calculate the MTF data of the lens module.The main research contents and innovative methods are as follows:(1)The research background is summarized.Various methods of judging the quality of the imaging system are investigated.The significance of measuring the MTF of the optical imaging system is clarified.The domestic and foreign research trends and developments of the modulation transfer function are investigated and analyzed.And the research goals of the thesis are proposed.(2)The basic principles and calculation methods of modulation transfer function are researched and analyzed.The pros and cons of various methods in the experiment are compared,and the improved edge method is selected to calculate the MTF data of the lens module,which can be possible to conduct objective,fast and reliable quality assessment of the lens module of the packaged equipment.(3)The traditional slanted edge method is analyzed and an improved slanted edge method is proposed to obtain MTF.Canny operator combined with cubic function fitting method is used to track the edge,obtain accurate sub-pixel edge points,and the five-point linear filtering method is used to denoise the edge spread function(ESF)sample points after oversampling.Nonlinear least square method is used to solve the parameters of the improved Hyperbolic function,and function fitting on the ESF sample points is performed,which can effectively reduce noise impact on MTF data,and improve data accuracy.Finally,simulation and experiment are performed on the improved edge method proposed to verify its feasibility and accuracy.(4)Focus on researching MTF test platform and host computer software production,the construction of test platform is studied,including selection and adjustment of light source,brightness and image transmission mode.The software interface is programmed in Visual C++ language,to cooperate with the open source library of industrial cameras to realize real-time image acquisition,camera parameter setting,basic image processing and MTF data calculation and other functions.The software is with user friendly interface and is easy to use.The test platform is introduced in detail,and the experimental operation steps are explained to ensure the MTF test of the lens module in consistent conditions.(5)After using the principle of optical application to make fine adjustments the light source and lens inside the lens module,the MTF measurement of the lens module is experimentally verified to prove the correctness and stability of the algorithm and software operation.The influence of the test results,through the modulation transfer function area(MTFA)of the lens module at a specific frequency as the reference index,the quality inspection and evaluation of the lens module on the production line are carried out.The experimental results show that the proposed method has better test accuracy and stronger noise immunity.