The Study Of Image Stitching System With Large Field Of View And High Resolution

In the night and bad weather conditions, it is a urgent problem need to be solved to protect automobile driving safety, if the vision scope of the driver and automobile lighting system can be actively and effectively extended, the traffic accidents can be greatly reduced, it can be shown that the automotive night vision system can be satisfied the target. The system index was determined in the paper firstly, and the design of the system hardware was finished. In order to satisfy the system real-time requirement, the dual processor framework FPGA+DSP was adopted. Firstly, the image stitching algorithm was studied, and Particle Swarm Optimization(PSO) was improved, and the MPSO was proposed and was used to fulfill the image registration. The partial image algorithm and software design based on the FPGA were mainly studied about system processor, the image capture, image storage and image prior process algorithm was fulfilled. It can be seen from the simulation experiment that the accurate image registration quality can be acquired by MPSO algorithm, at the same time the real-time can be satisfied, so the MPSO algorithm is a robustness algorithm. It also can be seen that the image capture, image storage and image prior process can be correctly fulfilled in the experiments base on FPGA. The processed signal outcome was stored into SDRAM and was processed by DSP in the next stage.