Research And Development On Embedded Machine Vision System

Machine vision system is applied to the fields that include industrial production, military and medicine widely. The mainstream machine vision system, which is constituted by CCD camera, image acquisition card and computer, has some disadvantages of complicated structure, high-cost, large cubage and so on. In recent years, with the incessant development of sensor, semiconductor and embedded system technology, a new machine vision system, namely embedded machine vision system stepped into course of development mileage of machine vision system phylogeny. Wherewith the advantage of high-performance processing ability, reliability, small cubage, low power, low cost and good-security, embedded machine vision system, which has extensive prospects, can conquer the disadvantage of conventional machine vision system effectually.The design of this system is a test research of incorporation of embedded system and machine vision, and it also provides a specific scheme of embedded machine vision system developing. To implement image acquisition in high speed, the CMOS sensor is directly controlled by ARM which is embedded processor with 32 bit. This paper composed of hardware designing and software designing. In the hardware designing, the writer chose the core device of the subsystem and provided the general block diagram after analyzing the function of each subsystem, and then designed the power module, the connection module of CMOS sensor and S3C2410, SCCB bus interface module and the module of image data storing. In the software designing, the writer wrote the bootload program, RS232 drive program, SCCB bus drive program and the program of image data transmission in DMA mode after analyzing the operating principle of each module device. At last, the paper gives a specific research on how to reduce the power consumption, and gives a specific method after that.As an intellectualized detection system, it is the result of the development of optics, mechanics and computer science, and it also merges embedded technology, machine vision technology, image acquisition and image processing technology. Sothere is not only abundant theory researching value but also widespread application value on this system.