The Study of Inspecting LED Chip by Image Processing and Piezo Servo System

• Main Authors: Kai-Chu Chuang, 莊凱竹
• Other Authors: Yi-Cheng Huang
• Format: Others
• Language: zh-TW
• Published: 2006
• Online Access: http://ndltd.ncl.edu.tw/handle/38703363161802661721

碩士 === 國立彰化師範大學 === 機電工程學系 === 94 === The aim of this paper is to develop an innovative inspection system for present emerging LED (Light Emitting Diode) photoelectric industry. To cope with the processing size of next generation dwindles, the gear of the inspection equipment in ultra precision and fast inspection specifications will become one of the important research keys. This thesis develops a machine vision system to check LED micro-structure by embedded piezo probing stage. Vision servo is used for feedback control on positioning. Piezo-actuated probing stage is controlled by using feed-forward neural network (FFNN) learning algorithm to compensate the nonlinear and hysteresis properties of piezoelectricity. Fast finished lighting LED is done thereafter. Experimental results show each LED Pad/Die can be finished lighting within the rate of 4 pieces per second of one probe and validates a new LED inspection system. Keywords：LED, Machine Vision, Piezoelectric Actuator, Feed-Forward Neural Network