Based on the idea of subspace projection, an optimal kernel discriminant analysis (OKDA) is given and used to extract features from a range profile.
英
美
- 核判别分析(KDA)是一种抽取非线性特征的有效方法,但它会因为奇异性问题而难以求解。