关键词: 机器视觉, 现场可编程门阵列, 数字信号处理, 异纤检测

Abstract: Foreign fiber of only a small proportion of cotton causes great harm.An improved Canny edge detection algorithm and a RGB threshold detection algorithm based on region grow algorithm are proposed，and a high-speed foreign fiber detecting and clearing system based on machine vision is developed，as there are some questions about current foreign fiber detecting system such as low efficiency，poor stability etc.A high-speed image acquisition and processing card based on FPGA and DSP is designed to detect foreign fiber.The foreign fiber detecting system has double layers illumination and PLC on-line monitoring module.The experimental results show the validity and sophistication of both algorithms for foreign fiber detecting，and the proposed system not only exhibits better robust stability but also enhances foreign fiber detecting speed and accuracy.

Key words: machine vision, Field Programmable Gate Array（FPGA）, Digital Signal Processing（DSP）, foreign fiber detecting