Title | ||
---|---|---|
A Novel Hybrid Algorithm of Sea Object Classification Based on Multi-sensor and Multi-level Track |
Abstract | ||
---|---|---|
To classify sea targets of underwater and surface groups. A novel hybrid classification algorithm based on sonar, automatic identification system (AIS) and radar is proposed in this paper. The proposed method includes four parts. The data preprocessing, the multi-target data association, the multi-sensor multi-target correlation, and the underwater/surface probability distribution fusion. Firstly, the measurement data of multiple sensors are unified in time and space through space-time registration. Secondly, the measurement data of each sensor are separated into different target sets by Mahalanobis distance discriminant method. And each target is modeled by grey prediction GM (1,1) model subsequently, and the noise of data are filtered by Kalman filter (KF). Thirdly, it preliminarily determines the type of targets by Hungarian algorithm. Finally, the D–S evidence theory based on the Angle cosine and Lance distance (ALDS) is used to further determines the target type. The proposed methods can be applied when there is inconsistent evidence. Simulation results illustrate that the proposed algorithm is effective in decision support for sea target classification. |
Year | DOI | Venue |
---|---|---|
2022 | 10.1007/s40815-022-01252-9 | International Journal of Fuzzy Systems |
Keywords | DocType | Volume |
Grey prediction model, D–S evidence theory, Hungarian algorithm, Kalman filter, Mahalanobis distance discrimination | Journal | 24 |
Issue | ISSN | Citations |
6 | 1562-2479 | 0 |
PageRank | References | Authors |
0.34 | 14 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Daqi Zhu | 1 | 25 | 5.09 |
Zhenzhen Zhang | 2 | 0 | 0.34 |
Mingzhong Yan | 3 | 0 | 0.34 |