| Autor: |
Xiaohui Bai, Shuwen Xu, Jianan Zhu, Zixun Guo, Penglang Shui |
| Jazyk: |
angličtina |
| Rok vydání: |
2023 |
| Předmět: |
|
| Zdroj: |
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, Vol 16, Pp 9422-9436 (2023) |
| Druh dokumentu: |
article |
| ISSN: |
2151-1535 |
| DOI: |
10.1109/JSTARS.2023.3321998 |
| Popis: |
Amid radar target detection fields, the detection of small floating targets in sea clutter is a tough but crucial task. An efficient solution to address the issue is feature-based detection method based on distinguishable features extracted from sea clutter and returns with target. Nevertheless, as the number of features increases, traditional algorithms with dimension limitation are not adaptable to combine the multiple features to detect. In this article, angle variance-based anomaly detection algorithms which solve the problem of the “curse of dimensionality” are introduced into floating small target detection. An improved algorithm by using the Mahalanobis distance is proposed and for the purpose of detecting the sea-surface small target, a novel detector with controlled false alarm capability is developed. On the open and recognized databases, the detection abilities of the proposed detector are assessed. It is confirmed that the proposed multifeature angle variance-based detector is more effective than other detectors based on multifeature distance or density from experimental results. This proposed detector can implement the improvement of detection performance with less computational cost compared to other well-established detectors. |
| Databáze: |
Directory of Open Access Journals |
| Externí odkaz: |
|
