基于分块加权的LBP算法在人脸识别中的应用

Application of LBP Algorithm Based on Block Weighting in Face Recognition

目前,高智能化的人脸识别系统一般都是基于PC平台,系统复杂集成度高,需要优异的软硬件支持。嵌入式技术的应用,有利于提高人脸识别系统的灵活性与有效性。本文针对传统的人脸识别算法提出一种改进的LBP算法用于人脸识别。实验结果表明:该算法在ORL、Yale及自建的人脸库上也能进行人脸识别。

At present, the highly intelligent face recognition system is generally based on the PC platform. The system is highly complex and requires excellent hardware and software support. The application of embedded technology is conducive to improving the flexibility and effectiveness of face recognition system. In this paper, we proposed an improved LBP algorithm for face recognition based on the traditional face recognition algorithm. Experimental results showed that the algorithm could recognize faces in ORL, Yale and self built face database.