红外凝视成像系统弱点目标检测的实时实现

Real Time Realization on DSP for Dim Target Detection in Staring Imaging System

为了解决红外点目标检测算法的复杂性与系统实时性之间的矛盾,设计了一套基于DSP的实时处理硬件实施方案。首先,介绍了基于高性能DSP的实时处理系统的原理结构,在此基础上,从硬件和软件两方面提出了系统实时实现的关键技术,包括大容量图像数据的双缓冲交互访问机制、图像预处理算法的FPGA快速实现,以及基于多媒体指令集等技术的代码级优化策略。实践结果证明,优化后的实时处理实现方案有效地提高了系统的实时性和稳定性,能够满足红外点目标检测算法大数据量高速传输、复杂运算的实际需求。

In view of the conflict between algorithm＇s complexity and the real-time property in the infrared dim target detection system, a high speed DSP operation project is designed. Firstly, the hardware principle frame based on high performance DSP is introduced, then the key technologies of real-time implementation are proposed from hardware and software aspects, which include the mechanism of mass image data double caching, the fast implementation of image preprocessing with FPGA and code optimizing based on multimedia instructions. The results show that the speed and stability of the system have been remarkably improved, which could meet the requirements of mass data high-speed transmission and complex operation of infrared dim target detection and tracking system.