基于DM642的实时人脸检测系统的实现  

Real Time Implementation of the Face-Detection System Based on DM642

作　　者：李月芳[1] 

机构地区：[1]常州轻工职业技术学院电子电气工程系,江苏常州213164

出　　处：《电子器件》2014年第3期454-458,共5页Chinese Journal of Electron Devices

摘　　要：人脸检测作为人脸分析的首要环节,在计算机视觉、模式识别、多媒体技术研究等领域中占有日益重要的地位。为了在嵌入式系统中实现人脸的实时检测,研究了基于TMS320DM642的人脸检测设计方案。提出了一种基于简化Gabor小波特征和层次型支持向量机的人脸检测算法,针对该算法与TMS320DM642的结合进行了系统设计和代码移植,阐述了算法性能的优化方案。优化后的检测算法满足了实时性要求,为实际应用奠定了基础。As the first and the primary procedure of the human-face-analysis, face-detection plays an increasingly important role in the areas of computer vision, pattern recognition and multimedia. In order to realize the real-time face-detection in the embedded systems, the face-detection scheme based on TDS320DM642 is investigated. A face- detection algorithm, based on the simplified Gabor wavelets and the hierarchical support vector machines, is proposed. The system design, code porting and performance optimization scheme in the implementation of the algorithm based on TDS320DM642 are detailed. The proposed algorithm satisfies the real time requirements, and lays the foundation for the future practical application.

关 键 词：人脸检测 TDS320DM642 GABOR小波 支持向量机 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]