Study and Application of Image Recognition on Industrial Control Platform

• Main Authors: CHIH-SHAN HUNG, 洪執善
• Other Authors: Chou, Yun-hsih
• Format: Others
• Language: zh-TW
• Published: 2016
• Online Access: http://ndltd.ncl.edu.tw/handle/24594979963558610558

碩士 === 聖約翰科技大學 === 電子工程系碩士班 === 104 === In recent year, automatic technology are booming to reduce the labor costs of production and maintenance, robots and many kind of automated production equipment are widely used by enterprises. We use PLC controller with stepper motor and motor controller to establish a platform of industrial automation, and using a fixed position camera on this platform to recognize screws. In this paper, we proposed a method of blind recognition algorithm which can recognize screws without pre-learning or database. In order to increase the correct rate of recognition, we use erosion and expansion method to reduce the noise in the origin image, and we discuss the shape and size of the mask which is necessary for image pre-processing. After processing the image, we will capture most likely screw element to achieve the goal of screws blind recognition.