摘要
One obvious trend for scholars is seeking an appropriate data description in hyperspectral anomaly detection. However, a specific predetermined model may not be able to fit all the other cases of hyperspectral images (HSIs). Hence, in this paper, we propose a graphical score estimation based hyperspectral anomaly detector (GSEAD) that utilizes a graphical data description of the HSI to achieve a data-adaptive analysis based anomaly detection procedure. Firstly, potential anomalies are screened out by a predicted connected component graph (pcc-graph). The remaining pixels constitute the robust background dataset. Secondly, an embedded locality preserving graph (elp-graph) is generated with the robust background dataset in an intrinsic manifold space by locality preserving graph embedding. Finally, a k-nearest neighbor (k-NN) score estimation is undertaken on the basis of the elp-graph. The experimental results confirm that the proposed method outperforms some of the state-of-the-art anomaly detectors.
| 源语言 | 英语 |
|---|---|
| 页数 | 5 |
| 出版状态 | 已出版 - 21 8月 2016 |
| 活动 | 8th Workshop on Hyperspectral Image and Signal Processing - Evolution in Remote Sensing (WHISPERS) - Los Angeles,CA, Los Angeles, 加拿大 期限: 21 8月 2016 → 24 8月 2016 |
会议
| 会议 | 8th Workshop on Hyperspectral Image and Signal Processing - Evolution in Remote Sensing (WHISPERS) |
|---|---|
| 国家/地区 | 加拿大 |
| 市 | Los Angeles |
| 时期 | 21/08/16 → 24/08/16 |
关键词
- Hyperspectral
- graphical data description
- score estimation
- anomaly detection
收录
- CPCI-S
学术指纹
探究 'GSEAD: Graphical Score Estimation for Hyperspectral Anomaly Detection' 的科研主题。它们共同构成独一无二的学术指纹。引用此
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver