Research On Optical Imaging System Of ACF Partical In Touch Screen

With the development of science and technology,the market requirements for quality of touch screen production are also getting higher and higher,and the lamination status of ACF conductive particles on COG area determines the quality directly,and it will cause poor conductivity if the number of ACF particles is too small,the Indentation is too light or the distribution is uneven.At present the detection of ACF particles rely on manual detection in domestic through microscope.This approach is subjective,time-consuming and laborious,and has high false rate and leak detection rate,and this greatly limits the development of the domestic touch screen industry,so the automatic optical inspection of ACF particles is proposed to meet the needs of the industry.As a key part of automatic optical inspection,the design of optical imaging system directly determines the quality of the test results are good or bad.In this dissertation,by studying the basic information of ACF particles,a set of optical imaging systems that meet the needs of the industry detection has been designed,and a series of experiments have been done to adjust the experimental parameters,and then improve the system.A set of optical imaging module has been designed based on touch screen manufacturing process,particle size and other information in the first place in this dissertation.Compared with traditional imaging components,DIC prisms are added to change particle area grayscale information to capture particle image,and a line-array camera with TDI mode is used to improve imaging resolution and sensitivity to get more clearly micron-sized particle image,and coaxial point light source with stable operation and easy installation is used to facilitate experimental research and industrial installation.Then the corresponding micron-class high-precision movement module is assembled according to the imaging module that consist of X,Y and Z three directions of the displacement platform and the adsorption stage to achieve accurately control of the speed and position of the scanning motion to ensure the smoothness and accuracy of the movement.Then five image quality evaluation methods have been proposed to detect the image changes of different parameters.By comparing the applicability of different evaluation methods,it is determined that the image quality captured by the optical imaging system under each parameter can be judged from the five aspects of absolute subjective evaluation,PSNR and MSE,information entropy,Brenner function,and gray range contrast in the particle area.Then,the effect of camera TDI series and wavelength of illumination source on image quality has been experimentally studied.Setting the camera TDI series at 4,8,16,32 levels,collecting the particle image under white light illumination and evaluating the image quality.According to the evaluation results,particle image has the best quality at 32 TDI level.And then,the red,yellow,green and blue light are used as the system illumination source.Capturing particle image under the 32 level TDI and analyzing the image quality.According to the results,green light is the most suitable illumination source.Finally,the optical imaging system has been installed in an industrial detection device.The particle image collected by the imaging system can be accurately marked by the image processing system of the detection device.That is,the imaging system is successfully applied to industrial automation testing.The characteristic of this dissertation is to propose a suitable optical imaging system that can capture high quality particle images for automatic optical detection and can be successfully applied in industrial automatic detection.It can replace the traditional manual microscopy detection and achieve optical automatic detection.