Research On Laser Cutting System Of LCD Based On Machine Vision

The production process of LCD screen is very complex,requiring nearly 100 processes,so it is inevitable that there will be various types of defects in the production process.In order to improve the production efficiency and reduce the production cost,a number of test and repair links will be set up in the LCD production process.In each test and repair process,it is often necessary to light up the display panel through the display panel lighting system in order to find all kinds of defects of the LCD screen.And LCD screen before entering the module system need to light a link in the test circuit resection,to prevent circuit and IC device circuit concatenated form a short circuit,burn out the LCD panel,so the test circuit of cutting process is an important process in the process of LCD production,test circuit,good or bad will directly affect the cutting quality of the follow-up process.In view of the above requirements,the paper carried out research on the visual positioning technology of Mark point and visual detection technology of cutting line,and designed a laser cutting system guided by vision and backbeat detection of circuit cutting quality.The system is mainly composed of two modules: visual guidance and visual detection.The visual guidance module locates the multi-section cutting area by recognizing the Mark point on the LCD screen,so as to guide the laser to cut.The visual detection module is mainly used to detect the cutting line,and detect the problems such as missing cutting,cutting channel deviation,incomplete cutting and excessive cutting.The main research contents of this paper are as follows:（1）According to the project requirements,define the key technologies and Research and development goal of the project,complete the selection and scheme design of visual hardware,and determine the overall design of the system and communication scheme.（2）Three common visual positioning methods were selected for theoretical analysis and experimental comparative testing,and the advantages and disadvantages of the three methods were compared.Finally,the template matching positioning method was selected for Mark point positioning,and the corresponding image pre-processing algorithm was used to complete the positioning algorithm design.Then the camera calibration and the unification of the camera and laser coordinate system are introduced,and the two-point positioning algorithm designed to improve the machining efficiency is introduced in detail.（3）through the analysis of the difficulties of cutting line positioning,in the aspect of software design using human-computer interaction way to determine the scope of the search,coupled with the regional morphological image processing method to complete the positioning of cutting line,finally based on the topological refinement measure method to extract the center line of the cutting line and makes the corresponding judgment about the quality of the cutting line.（4）Halcon visual algorithm library and VS2019 platform were used to complete the software development of laser tangent system,and the operating mode and workflow of the software were introduced.Due to the inevitable mechanical errors in the whole system,the software designed a series of compensation methods to make the whole system to reach the accuracy required by customers.Through the experimental test and field production,the software has reached the initial research and development goals and runs stably.