Study On The Auto Detect System For Displays Dysfunction Base On OSD Interaction

With the rapid development of electronic information technology,human beings entered the information age,and displays become important carrier of information transmission.In the industrial manufacturing,rigorous test is required for displays from product development to the market.Due to the growth of user demands,the display's OSD(On Screen Display)menu functions getting rich,and the subsequent testing tasks are also rising.In order to alleviate the pressure of the industrial test,the research of auto test technology for displays rearch boom in relevant institutions at home and abroad.In view of the current problem for the low degree of automation in displays function test and the test item is single,this paper designs an automatic detect system for displays dysfunction based on OSD interaction to alleviate the pressure of manual test and enhance the test efficiency.First,in terms of the test cases execution,this paper puts forward a control scheme of the interaction between PC and the display's OSD menu.By studying the architectures of the displays driver circuits,design a communication conversion circuit to establish the communication between PC and the display's data bus.Then define the menu's control instructions and burn to Scalar IC,further establish DDC/CI protocol to execute OSD interaction test cases by software,and complete the auto setting of display menu environment in the test process.Then in terms of the anomalies detection,divided the display's screen into video image area and the OSD menu area.According to the different forms of video image anomalies,detailing the color parameters abnormal and screen flicker abnormal.In view of color parameters abnormal,choose high precision color analyze instrument,and put forward color correction method,then complete automated optical data collection to analyze and detect the anomalies;in view of flicker detection,collect the light signal and then translate it into voltage waveform,build a flicker perception model to simulate the human eyes perception and evaluate the flicker.In terms of OSD menu abnormal detection,using the method of machine recognition to prove the validity of the character information in the display menu.First,take the menu photo by CCD camera,and do preprocessing including graying,binarization,noise reduction,tilt correction,etc,then use Tesseract engine to train OSD menu character and recognize the menu characters.Compared the results with the expected status so that can detect the abnormal state of the menu.Finally,using visual programming to integrate the modules and develop the software,then build the system's test environment to do actual test verification.The results show that the video image optical detect error is within 5%in the system,the average recognition accuracy reaches 95.5%for different OSD language mode,and the automatic and semi-automatic degree reache 69.2%and 19.6%respectively.According to TPV's benefit evaluation,the system may create 168300￥ benefits every year.