摘要
为解决电视捕获跟踪瞄准系统中算法复杂性与系统实时性之间的矛盾。本文设计了以最新DSP芯片TMS320C6202为核心处理器、基于FPGA和PCI总线的实时多目标识别跟踪处理平台。通过对C6202和C6201芯片性能的比较,得出C6202适合实时图像处理的特点。重点介绍了该实时处理平台的基本组成、工作原理和系统优化措施。针对复杂背景下目标多、小、弱的特点,初步探讨了一种易于FPGA硬件实现的模糊边缘检测算法的理论基础。
In order to resolve the confliction between the complexity of the algorithm and the real-time character in the TV capturing, tracking and collimation system, a real-time hardware platform is designed to identify and track many targets. The platform is based on FPGA and PCI Bus, and TMS320C6202 is its core processor. Compared with the characters of C6201 and C6202, the conclusion is that C6202 is suit for the real-time image processing. The basic frame, the working principle and the optimization ways of the system are introduced in detail. Based on the fact that the identified targets are many, small and weak in the complex background, the algorithm of fuzzy image border extracted is introduced and it can easily be realized with FPGA.
出处
《信号处理》
CSCD
2002年第2期155-158,146,共5页
Journal of Signal Processing