Vision System For Mini Unmanned Aerial Vehicles Based On DSP

Mini UAV (Unmanned Aerial Vehicle) which features low cost and high flexibility has a broad application in the civil and army field. As an indispensable part of Mini UAV, vision system has been paid great attention by colleges and research organizations in the world. A typical vision system is designed to implement air-shooting, object recognition, vision navigation and many other tasks. It is significant to make a research on vision system for Mini UAV.This paper focuses on how to establish an integrated vision system for Mini UAV from both the hardware and software aspects. The hardware part chose TMS320DM642 DSP as the core processing platform, and was augmented by the capture module, communication module, power supply module, etc. The software design realized the averaging filter, binarization, detection, skeletonization image pre-processing and Randomized Hough Transform image processing algorithms. The system could implement landing-target recognition and estimate the distance to Mini UAV on the basis of recognition result in surface layer simulating experiment.Compare to the traditional methods which features analog image transmission and PC processing platform used by domestic research organizations. The DSP vision system designed in this paper get rid of the restriction of link's precision for analog transmission, has the advantages of higher Cost-Performance Ratio, and is designed for combining with flight control system, shows the future of the related research area.