Real - Time Image Preprocessing Technology Based On

Nowadays, with the development of society, image processing has got a lot of attention. The applications of it has become more and more widely, such as pattern recognition, computer vision, medical image processing and other fields,in which image processing is needed very necessary. But with the development of science and technology, the resolution of the image is gradually increased.People’s requirement of real-time image processing also increases and the commonly used image processing method based on the software has been unable to meet the needs of real-time. FPGA has the characteristics of programmable, reconfiguration and parallel processing, which is slowly being used in the image processing system.This paper applied a image acquisition which is assembled by MT9M034 sensor and FPGA development board with EP3C16F484 chip belonging to three generation series. It can collect, store and display color images. In addition, this paper also puts forward a non-uniform illumination image segmentation method.The method combines Prewitt operator and Bernsen local binarization to segment image. The paper writes this method to the above the system which can segment the image in real-time by FPGA. The experimental result shows that the method can real-time segment image and have a good inhibiting effect on a non-uniform illumination image. Besides, this paper implements commonly used image processing algorithms on FPGA platform and then compares the time taken by these algorithms which are implemented on MATLAB and VC.The hardware platform has obvious advantage to real-time preprocess the high resolution image.