基于机器视觉的色差检测系统设计-控制理论与控制工程专业论文.docxVIP

华中科技大学硕士学位论文 华 中 科 技 大 学 硕 士 学 位 论 文 II II Abstract With the development of computer technology and digital imaging technology, machine vision technology has been applied in nearly all of domains that need human vision, such as science and technology, national defense, aviation and space navigation, medicine and therapies, and all fields of national economy. The development of the platform for embedding and imaging technology in industry, the reduction of their cost and the enhancement of their rea-ltime performance boost the rapid development in industrial vision system. The texile and dying industry is one of the most important industries which can affect the ecomony and people’s living of the country. The detection of shading variation of cloth is very critical, because it is a main sign of clothes’ quality. By making a brief description of the development and application of machine vision system in domestically and abroad, and in terms of the case of shading variation measurement in civil dyeworks, the thesis puts forward a shading variation system based on machine vision, which is structured in embedded system based on DSP and the platform of PC. The system solves the bandwidth problem in the PC and frame grabber structure and the limited data memory problem in embedded system, and satisfies the real-time requirement. For the step motor driver in the system, after the abundant experiments, the thesis brings forward a curve for the motor’s acceleration and deceleration. The curve is very successful, and it has good results. In the future, the system will reduce the labor cost and improve the quality and speed of coloring cloths, which will enhance the competitive capacity in the market. On the basis of the analysis of the image processor, illumination system, image sensor and so on, the thesis designs the formula for the shading variation measurement system based on machine vision and introduces the design detail on the hardware and software. Using pictures and