A novel feature extraction algorithm of acoustic targets based on locality preserving discriminant projections
Wang Yi, Junan Yang, Hui Liu · 2010
This paper proposes a new supervised manifold learning algorithm called locality preserving discriminant projections (LPDP) to solve the problem of poor robustness in acoustic targets recognition. The algorithm is based on locality preserving projections (LPP) and the method called modified maximum margin criterion (MMMC) which is adopted to explore the optimal linear transformation for translation and rescaling automatically. So the proposed algorithm can not only achieve good results in classification, but also solve the small sample size problem and has the ability of out-of-sample learning. Many experiments are carried out with public databases to test the algorithm. Experiment results show that the proposed algorithm is more precise and stable than others.