The Recognition System For The Blind Sidewalk Based On FPGA

Nowadays,blind sidewalks can be seen in almost all the cities. Because of these facilities, the blind can go outside much more easily. But in reality there are something going as we don’t expect. Some blind ways haven’t been paved in a standard rule, or because they were built too long ago, so they can’t be distinguished so accurately. And others are destroyed. These blind sidewalks not only have no help with the blind, but also can bring them a lot of danger. Therefore, the design of a blind sidewalk system has an important practical significance. The design based on FPGA(Field Programmable Gate Array) can make full use of the parallelism character of FPGA. At the same time, the SOPC technology can make the design more flexible.This thesis designs a blind sidewalk recognition system based on FPGA. The system uses Altera DE2 development board as the hardware platform. With a CMOS(Complementary Metal Oxide Semiconductor) image sensor and a VGA(Video Graphic Array) device, it can be used for debugging and maintenance. When direction of the blind sidewalk changes a lot or an obstacle occurs, the system will output voice information via a speaker.The whole system consists of four modules, which are image capture module, image cache module, image processing module and image display module. It adopts a synergistic manner through software and hardware. The image capture module, image cache module and image display module are written in Verilog hardware description language. The image processing module is designed by SOPC Builder and programmed in C language in Nios II IDE. The thesis discussed and analyzed the selection of algorithms in each part. Among them, the image processing realizes the main algorithms of the system. Firstly,we adopt a median filter to smooth the gray level image. Secondly we use a algorithm to convert the image into a binary one,and this algorithm base on the Sobel edge detection. Then we detect the direction of the blind ways by Hough transform. At last we calculate the sum of the value using the binary image, and figure out a danger factor. And we can know if there is an obstacle with that factor. After repeated debugging and improvement, the system can achieve the direction of the blind sidewalks and judge if there is an obstacle ahead.