摘要
非均匀非平稳非高斯强杂波常导致基于能量的传统阈值检测方法目标检测率低或产生大量虚警,影响目标录取性能。一些工作探索了采用时频特征图基于机器学习的目标检测方法,但对于驻留时间短的雷达搜索应用,难以获得高分辨率时频特征,导致基于时频特征图的机器学习方法性能下降甚至失效。因此,提出了一种新的基于目标帧间多维特征的目标检测方法及相应的神经网络模型RDF-ResNet。通过在解模糊过程中提取疑似目标帧间多维特征并输入到RDF-ResNet中,实现在特征空间上对虚警的抑制,结合低阈值检测,实现检测率的有效提升。实测实验数据表明:文中所提方法可实现较传统阈值检测方法约41%的检测率提升和约48%的虚警率降低,能有效提升雷达目标检测能力,并为雷达回波特征空间的有效构建和机器学习雷达目标检测提供了新思路。
Strong non-uniform, non-stationary and non-Gaussian clutter typically leads to low detection rate or high false-alarm rate for conventional threshold detection methods in energy domain. Target detection based on machine learning with the time-frequency features has been explored by some previous works. However, such features may be unavailable for short dwell-time in real searching applications, which results in the performance degradation or even failure for these methods. Therefore, a new radar target detection method with multi-frame multi-dimensional features based on a new deep neural network RDF-ResNet is proposed. By exploiting multi-frame multi-dimensional features of targets during ambiguity resolution, the proposed RDF-ResNet can suppress false alarms effectively. Together with low-threshold detection, a significant improvement of detection performance of radars can be achieved by the proposed detection method. Field experiment demonstrates that the proposed method can increase the detection rate by 41% and reduce the false-alarm rate by 48% compared to the traditional threshold detection. A new way is also provided to establish feature spaces for radar echoes and detect targets with machine learning more effectively.
作者
王治飞
于俊朋
杨予昊
夏凌昊
WANG Zhifei;YU Junpeng;YANG Yuhao;XIA Linghao(Nanjing Research Institute of Electronics Technology,Nanjing 210039,China;Key Laboratory of IntelliSense Technology,CETC,Nanjing 210039,China;Jiangsu Provincial Key Laboratory of Detection and Perception Technology,Nanjing 210039,China)
出处
《现代雷达》
CSCD
北大核心
2022年第12期48-54,共7页
Modern Radar
关键词
多维特征
深度学习
卷积神经网络
目标检测
multi-dimensional features
deep learning
residual convolution neural network
object detection
