A convolutional neural network based approach to sea clutter suppression for small boat detection

作     者：Guan-qing LI Zhi-yong SONG Qiang FU Guan-qing LI;Zhi-yong SONG;Qiang FU

作者机构：National Key Laboratory of Science and Technology on ATRCollege of Electronic Science and TechnologyNational University of Defense TechnologyChangsha 410073China 

出 版 物：《Frontiers of Information Technology & Electronic Engineering》 (信息与电子工程前沿（英文版）)

年 卷 期：2020年第21卷第10期

页      面：1504-1520页

核心收录：

学科分类：080904[工学-电磁场与微波技术] 0810[工学-信息与通信工程] 0809[工学-电子科学与技术（可授工学、理学学位）] 08[工学] 081105[工学-导航、制导与控制] 081001[工学-通信与信息系统] 081002[工学-信号与信息处理] 0825[工学-航空宇航科学与技术] 0811[工学-控制科学与工程] 

主　　题：Convolutional neural networks Class activation map Short-time Fourier transform Small target detection Sea clutter suppression 

摘      要：Current methods for radar target detection usually work on the basis of high signal-to-clutter *** this paper we propose a novel convolutional neural network based dual-activated clutter suppression algorithm,to solve the problem caused by low signal-to-clutter ratios in actual situations on the sea *** activation has two ***,we multiply the activated weights of the last dense layer with the activated feature maps from the upsample *** this,we can obtain the class activation maps(CAMs),which correspond to the positive region of the sea ***,we obtain the suppression coefficients by mapping the CAM inversely to the sea clutter ***,we obtain the activated range-Doppler maps by multiplying the coefficients with the raw range-Doppler *** addition,we propose a sampling-based data augmentation method and an effective multiclass coding method to improve the prediction *** on real datasets verified the effectiveness of the proposed method.