基于梯形跨尺度特征耦合网络的SAR图像舰船检测  

Ship Detection in SAR Images Based on Trapezoidal Cross-scale Feature-coupling Network

作　　者：黄帅 张毅[1] HUANG Shuai;ZHANG Yi(Aerospace Information Research Institute,Chinese Academy of Sciences,Beijing 100190,China;School of Electronics,Electrical and Communication Engineering,University of Chinese Academy of Sciences,Beijing 100049,China)

机构地区：[1]中国科学院空天信息研究院,北京1000190 [2]中国科学院大学电子电气与通信工程学院,北京100049

出　　处：《计算机工程》2022年第12期270-280,共11页Computer Engineering

基　　金：国家部委基金。

摘　　要：在合成孔径雷达(SAR)图像舰船检测中,现有检测方法难以有效提取多尺度语义信息,无法准确地表示其在整个网络中的信息权重,且定位模块与分类模块相关性较弱,导致定位不准确。提出一种梯形跨尺度特征耦合网络,通过梯形特征金字塔网络提取各级语义信息,采用交叉结构代替跳连结构,提高网络的泛化能力和语意表征能力,并引入可训练权重因子表示各级语义信息的重要性。在此基础上,将定位模块与分类模块通过耦合检测头增强两者之间的相关性,引入可变形卷积对最终的定位输出进行二次校准,从而提高检测精度。实验结果表明,与FasterRCNN、CascadeRCNN、RetinaNet等主流网络相比,该网络在SSDD数据集上的检测精度提高了2.74个百分点以上,具有良好的检测性能。在近岸复杂场景下,该网络能更有效地检测密集目标和多尺度目标,降低误检和漏检的概率。In ship detection in Synthetic Aperture Radar(SAR)images,effectively extracting multi-scale semantic information by using existing detection methods is difficult.Additionally,the information weight in the whole network cannot be accurately represented,and the correlation between the positioning module and the classification module is weak,leading to inaccurate positioning.This study presents a trapezoidal cross-scale feature-coupling network,in which the semantic information at all levels is extracted by employing a trapezoidal feature pyramid network.The cross structure is used to replace the jump structure to improve the generalization ability and semantic representation ability of the network.A trainable weight factor is introduced to represent the importance of semantic information at all levels.The correlation between the positioning module and the classification module can be enhanced by coupling the detection head,and a deformable convolution is introduced to calibrate the final positioning output and improve the detection accuracy.Experimental results show that compared with mainstream networks,such as the FasterRCNN,CascadeRCNN,and RetinaNet,the detection accuracy of the proposed network on the SSDD dataset is improved by more than 2.74 percentage points,demonstrating good detection performance.In the near-shore complex scene,the network can detect dense and multi-scale targets more effectively and reduce the probability of false or missed detections.

关 键 词：舰船检测 梯形特征金字塔 多尺度特征聚合 耦合网络 可训练权重因子 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]