SVD和DCT提取特征向量的方法在人脸识别中的比较
Comparison Between SVD and DCT Feature Extraction Methods in Face Recognition
-
摘要: 采用奇异值分解和2D离散余弦变换两种方法得到的特征,分别在嵌入式隐马尔可夫模型的人脸识别 中进行比较,得出奇异值分解比2D离散余弦变换在识别率上和时间复杂度上都较差的结果。通过对结果进行 分析得出,虽然奇异值分解有很多优良的特性,但是在模式识别中仅使用奇异值分解来提取特征并不是很好的 方法。Abstract: This paper compares SVD (singular value decomposition) with DCT (discrete cosine transform) in the face recognition, and concludes that SVD is less effective than DCT in recognition ratio and time cost. Experimental results show that SVD is an available feature extraction method, but is not a best one for pattern recognition
下载:
