引用本文: | 董云龙, 张兆祥, 刘宁波, 黄 勇, 丁 昊, 张梦雨. 雷达回波三特征联合海况分类方法[J]. 雷达科学与技术, 2023, 21(2): 189-198.[点击复制] |
DONG Yunlong, ZHANG Zhaoxiang, LIU Ningbo, HUANG Yong, DING Hao, ZHANG Mengyu. Sea State Classification Method Based on Three Features of Radar Echo[J]. Radar Science and Technology, 2023, 21(2): 189-198.[点击复制] |
|
摘要: |
海况作为雷达对海上目标探测的重要背景信息,对其灵活、快速和准确判断在对海探测雷达的目标精细化探测中具有重要作用。为此,本文提出了海杂波时频谱的均值函数脊累积量、标准差函数脊累积量、多普勒谱的区域峰值功率与噪声均值功率比三种差异性特征来区分高/低海况,并利用支持向量机(SVM)构造了三特征联合的高/低海况分类器,最后使用海军航空大学“雷达对海探测数据共享计划”数据集对所提分类器进行测试,结果表明所提方法在使用64个相参脉冲的条件下,即可实现对高/低海况的准确分类,能够满足对海探测雷达工作于扫描模式的需求。 |
关键词: 对海雷达 海况分类 时频域特征 频域特征 支持向量机 |
DOI:DOI:10.3969/j.issn.1672-2337.2023.02.010 |
分类号:TN959.72 |
基金项目:国家自然科学基金(No.61871392, 62101583, 61871391); 山东省自然科学基金(No.ZR2021YQ43) |
|
Sea State Classification Method Based on Three Features of Radar Echo |
DONG Yunlong, ZHANG Zhaoxiang, LIU Ningbo, HUANG Yong, DING Hao, ZHANG Mengyu
|
1. Institute of Information Fusion, Naval Aviation University, Yantai 264001, China;2. Yantai University,Yantai 264001, China
|
Abstract: |
As an important background information for radar to detect floating targets on the sea, the flexible, fast and accurate judgment of sea state plays an important role in the target fine detection of sea detection radar. Therefore, this paper proposes three different features of the ridge integration of the mean function of the time?frequency spectrum, ridge integration of the standard deviation function and the ratio of regional peak power to noise mean power of Doppler power spectrum to distinguish high/low sea states, and designs a three?features combined high/low sea state classifier by using support vector machine (SVM). Finally, the proposed classifier is tested with the data set of “Radar?to?Sea Exploration Data Sharing Program” of Naval Aviation University. The results show that the proposed method can accurately classify the high/low sea state under the condition of using 64 coherent pulses, and can meet the requirements of general sea detection radar working in scanning mode. |
Key words: sea radar classification of sea state time⁃frequency domain features frequency domain features support vector machine |