Name
Title | ||
|---|---|---|
Symmetric Information-Theoretic Metric Learning for Target Detection in Hyperspectral Imagery |
Abstract | ||
|---|---|---|
Metric learning-based methods, which yield great performance and show considerable potential to improve the performance of hyperspectral image processing, aim to calculate the Mahalanobis distance metric matrix. In this article, we proposed a symmetric information-theoretic metric learning (SITML) method for hyperspectral target detection. The SITML algorithm is designed based on the classical information-theoretic metric learning (ITML) and, minimizes the differential Kullback-Leibler (KL) divergence. To enhance both of the detection performance and the generalization ability, we build metric spaces from the neighborhood of training samples to preserve the local discriminative information. Then, we conduct local pairwise constraints to maximize the Jeffery divergence (also named the symmetric KL divergence) of two multivariate Gaussian distributions to solve the problem of an asymmetric KL divergence. Finally, we use a closed-form solution to solve the optimization problem. Intensive experiments on three hyperspectral datasets indicate that SITML outperforms the classical ITML algorithm and other state-of-the-art target detection methods. |
Year | DOI | Venue |
|---|---|---|
2022 | 10.1109/JSTARS.2022.3140756 | IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING |
Keywords | DocType | Volume |
Measurement, Object detection, Hyperspectral imaging, Learning systems, Training, Extraterrestrial measurements, Symmetric matrices, Hyperspectral target detection, metric learning, symmetric Kullback-Leibler (KL) divergence | Journal | 15 |
ISSN | Citations | PageRank |
1939-1404 | 0 | 0.34 |
References | Authors | |
0 | 3 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Yanni Dong | 1 | 1 | 2.04 |
| Yuxiang Zhang | 2 | 11 | 15.58 |
| Bo Du | 3 | 1662 | 130.01 |