联合多尺度特征和注意力机制的遥感影像飞机目标检测  

Aircraft Target Detection in Remote Sensing Image Based on Multi-Scale and Attention Mechanism

作　　者：徐佰祺 江刚武[1] 刘建辉[1] 王鑫[1] 魏祥坡 余培东 XU Baiqi;JIANG Gangwu;LIU Jianhui;WANG Xin;WEI Xiangpo;YU Peidong(Information Engineering University,Zhengzhou 450001,China)

机构地区：[1]信息工程大学,河南郑州450001

出　　处：《测绘科学技术学报》2020年第4期398-403,共6页Journal of Geomatics Science and Technology

基　　金：国家自然科学基金项目(41801388)。

摘　　要：针对遥感影像飞机目标尺寸小、特征不明显的问题,在YOLO V4的基础上,提出一种联合多尺度特征和注意力机制的遥感影像飞机目标检测方法。该方法扩大了特征融合时尺度的范围,增强了对低层特征和小目标信息的提取。通过引入注意力机制进行特征融合,为每个通道的特征赋予不同权重,学习不同通道间特征的相关性。在RSOD-Dataset数据集上进行对比实验,实验结果表明该方法与相关算法相比,具有更高的检测精度。Aimed at the small and indistinguishable aircraft target in remote sensing image,a method of aircraft detection in remote sensing image is proposed based on YOLO V4.The scale of the feature fusion is expanded,features in lower layers and information of small size target are learned more by the algorithm.Then the attention mechanism is introduced to combine the features of each channel,which gives different weights to learn the correlation between different channels.Experimental evaluation is conducted on RSOD-Dataset.The results show that the method proposed in the paper has higher detection accuracy compared with other algorithms.

关 键 词：遥感影像 飞机目标检测 特征融合 注意力机制 YOLO V4算法 

分 类 号：P237[天文地球—摄影测量与遥感]