摘要
提出了一种利用Manhattan距离进行人脸表情分类的新方法。Manhattan距离计算出具有不同模式的两个对象的距离更大。在实验中,比较了Manhattan距离、欧氏距离、余弦距离在人脸表情分类中的性能,得出Manhattan距离比另外两类距离有着更好的识别效果。
The paper is presented a new method of facial expression classification.It's called Manhattan distance.Manhattan distance yields a higher value for pairs of objects that are less similar to one another.The author compares Manhattan distance with Euclidean distance and COS distance in the experiment.The performance of Manhattan distance is better than other methods.
出处
《计算机工程与应用》
CSCD
北大核心
2008年第2期74-75,79,共3页
Computer Engineering and Applications