Abstract | ||
---|---|---|
For feature extraction, the dimensionality of the feature space is usually much larger than the size of training set. This is known as under sample problem. At this time, local structure is more important than global structure. In this paper, locality sensitive batch feature extraction (LSBFE) is derived based on a new gradient optimization model by exploiting both local and global discriminant structure of data manifold. With the proposed LSBFE, a set of features can be extracted simultaneously. Recognition rate is improved compared with batch feature extraction (BFE), which only considers global information. It is shown that the proposed method achieves good performance for face databases, handwritten digit database, object database and DBWorld data set. |
Year | DOI | Venue |
---|---|---|
2016 | 10.1016/j.neucom.2015.07.076 | Neurocomputing |
Keywords | Field | DocType |
Feature extraction,Dimensionality reduction,Classification,Local learning,Under sample problem | Data mining,Dimensionality reduction,Computer science,Artificial intelligence,Kanade–Lucas–Tomasi feature tracker,k-nearest neighbors algorithm,Clustering high-dimensional data,Feature vector,Pattern recognition,Feature (computer vision),Feature extraction,Curse of dimensionality,Machine learning | Journal |
Volume | ISSN | Citations |
171 | 0925-2312 | 3 |
PageRank | References | Authors |
0.38 | 22 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Jie Ding | 1 | 10 | 3.56 |
Changyun Wen | 2 | 3686 | 284.86 |
Guoqi Li | 3 | 387 | 46.18 |
Chin-Seng Chua | 4 | 677 | 58.85 |