Target Detection in Sea Clutter Using a Three-feature Prediction-based Method
| Title: | Target Detection in Sea Clutter Using a Three-feature Prediction-based Method |
|---|---|
| Authors: | Yunlong DONG, Zhaoxiang ZHANG, Hao DING, Yong HUANG, Ningbo LIU |
| Source: | Leida xuebao, Vol 12, Iss 4, Pp 762-775 (2023) |
| Publisher Information: | China Science Publishing & Media Ltd. (CSPM), 2023. |
| Publication Year: | 2023 |
| Collection: | LCC:Electricity and magnetism |
| Subject Terms: | target detection, sea clutter, historical frame features, prior information, feature prediction, Electricity and magnetism, QC501-766 |
| More Details: | Feature-based detection methods are often employed to address the challenges related to small-target detection in sea clutter. These methods determine the presence or absence of a target based on whether the feature value falls within a certain judgment region. However, such methods often overlook the temporal information between features. In fact, the temporal correlation between historical and current frame data can provide valuable a priori information, thereby enabling the calculation of the feature value of the current frame. To this end, this paper proposes a novel method for time-series modeling and prediction of radar echoes using an Auto-Regressive (AR) model in the feature domain, leveraging a priori information from historical frame features. To verify the feasibility of AR modeling and prediction of feature sequences, the AR model was first employed in the modeling and 1-step prediction analysis of Average Amplitude (AA), Relative Doppler Peak Height (RDPH), and Frequency Peak-to-Average Ratio (FPAR) feature sequences. Next, a technique for extracting feature values by utilizing the temporal information of historical frame features as a priori information was proposed. Based on this approach, a small-target detection method predicated on three-feature prediction, which can effectively utilize the temporal information of historical frame features for AA, RDPH, and FPAR, was proposed. Finally, the validity of the proposed method was verified using a measured data set. |
| Document Type: | article |
| File Description: | electronic resource |
| Language: | English Chinese |
| ISSN: | 2095-283X |
| Relation: | https://doaj.org/toc/2095-283X |
| DOI: | 10.12000/JR23037 |
| Access URL: | https://doaj.org/article/446a4d06a8fc4d70beafefcf5f4a8c90 |
| Accession Number: | edsdoj.446a4d06a8fc4d70beafefcf5f4a8c90 |
| Database: | Directory of Open Access Journals |
| ISSN: | 2095283X |
|---|---|
| DOI: | 10.12000/JR23037 |
| Published in: | Leida xuebao |
| Language: | English Chinese |