Flexible non-greedy discriminant subspace feature extraction.
basic_science · Level V
Where this comes from
- Record sourced from PubMed, PMID 31063926.
- Also identified by DOI 10.1016/j.neunet.2019.04.006.
- No licence information is recorded for this record.
- Because redistribution is not established, this page shows the abstract only. Follow the links below for the full text.
Abstract
Recently, L<sub>1</sub>-norm-based non-greedy linear discriminant analysis (NLDA-L<sub>1</sub>) for feature extraction has been shown to be effective for dimensionality reduction, which obtains projection vectors by a non-greedy algorithm. However, it usually acquires unsatisfactory performances due to the utilization of L<sub>1</sub>-norm distance measurement. Therefore, in this brief paper, we propose a flexible non-greedy discriminant subspace feature extraction method, which is an extension of NLDA-L<sub>1</sub> by maximizing the ratio of L<sub>p</sub>-norm inter-class dispersion to intra-class dispersion. Besides, we put forward a powerful iterative algorithm to solve the resulted objective function and also conduct theoretical analysis on the algorithm. Finally, experimental results on image databases show the effectiveness of our method.
Medical subject headings
- Algorithms
- Pattern Recognition, Automated