Liquid Crystal Panel IC Area Conductive Particles Automatic Optics Detection Algorithm Research

In the industrial environment,the detection of liquid crystal panel is mostly manual inspection.The test result is subjective,the detection efficiency is low,and it is only a sampling test,which can not meet the requirements of industrial production on the quality of LCD panel.With the development of automation technology,automatic detection of LCD panel is in increasingly high demand.Automation technology is also the focus of research in various industries.In this thesis,the automatic optical detection method for the IC area of the liquid crystal panel can solve the problem of manual detection.The detection efficiency and false detection rate can meet the requirements of industrial production.Firstly,this thesis introduces the principle of COG technology.Based on the analysis of the disadvantages of manual detection,an automatic optical detection method based on machine vision is proposed.This method can make up for the shortage of manual testing and has low cost and high detection efficiency..Secondly,the overall design of the detection system is introduced,and the whole detection system is composed of two parts,hardware and software.It is divided into motion control module,image acquisition module,software interaction module and detection algorithm module.Among them,PLC controls the mechanical movement of the whole detection system,and the industrial personal computer controls the camera to acquire the image and runs the detection algorithm.Then,the collected conductive particles are analyzed.The original collection has a lot of information that is not interesting,so it is necessary to deal with the original image,extract the conductive particle area of interest from the whole image,denoise image and deal with uneven illumination.Finally,a particle detection algorithm based on the local mean difference of illumination direction is proposed.Through the enhancement of the gray characteristics of conductive particles,the false detection rate and the undetected rate of the conductive particles are greatly reduced.In view of the fact that a number of conductive particles may adhere to each other to form a large connected region,the local maximum method is used to calculate the number and location of the conductive particles.In view of the fact that the offset error caused by COG hot pressing can also lead to the failure of the touch panel,a method for calculating the offset value of the conductive particle binding region is proposed.The characteristic of this thesis is that it proposes a new algorithm for the detection of conductive particles.Compared with the extremum difference algorithm,the false rate is reduced by 1.3% and the detection rate is reduced by about 1.2%.In the double platform and LCD panel on working condition of 4 IC regions,to detect a plurality of LCD panels,LCD panel of the average running time of the algorithm is 2.51 seconds,the total system detection time is 4.85 seconds,fully meet the needs of the industrial test.The use of this method for the detection of conductive particles has not been reported in China.