关键词: 机器视觉, 自动识别, 序列号, 军用集成电路

Abstract: The existing technology needs to read the circuit serial number manually when testing the military integrated circuit serial number. The extraction of the test serial number needs to be correctly identified by naked eye and input into the test machine, which is error-prone and inefficient. The structure of military integrated circuit is mostly symmetrical distribution, and it can be placed into the socket regardless of forward or reverse placement. It requires manual naked eye identification and verification of the placement direction of the circuit, which has the risk of reverse discharge. Once the wrong direction (put), it is possible to cause integrated circuit damage or even scrap. Manual identification of serial number is time-consuming and laborious, which seriously affects the test output efficiency. Machine vision technology can be used to simulate human visual function, extract information from images of objective things, process and understand it, and finally use it for actual detection, measurement and control. It can not only avoid different detection results, reduce error, but also improve efficiency and accuracy.

Key words: machinevision, automaticidentification, serialnumber, militaryintegratedcircuit