融入混合注意力的可变形空洞卷积近岸SAR小舰船检测  

Deformable atrous convolution nearshore SAR small ship detection incorporating mixed attention

作　　者：龚声蓉 徐少杰[1,2] 周立凡 朱杰[1] 钟珊 Gong Shengrong;Xu Shaojie;Zhou Lifan;Zhu Jie;Zhong Shan(School of Computer and Information Technology,Northeast Petroleum University,Daqing 163318,China;School of Computer Science and Engineering,Changshu Institute of Technology,Changshu 215500,China)

机构地区：[1]东北石油大学计算机与信息技术学院,大庆163318 [2]常熟理工学院计算机科学与工程学院,常熟215500

出　　处：《中国图象图形学报》2022年第12期3663-3676,共14页Journal of Image and Graphics

基　　金：国家自然科学基金项目(61972059,42071438);江苏省自然科学基金项目(BK20191474)。

摘　　要：目的在近岸合成孔径雷达(synthetic aperture radar,SAR)图像舰船检测中,由于陆地建筑及岛屿等复杂背景的影响,小型舰船与周边相似建筑及岛屿容易混淆。现有方法通常使用固定大小的方形卷积核提取图像特征。但是小型舰船在图像中占比较小,且呈长条形倾斜分布。固定大小的方形卷积核引入了过多背景信息,对分类造成干扰。为此,本文针对SAR图像舰船目标提出一种基于可变形空洞卷积的骨干网络。方法首先用可变形空洞卷积核代替传统卷积核,使提取特征位置更贴合目标形状,强化对舰船目标本身区域和边缘特征的提取能力,减少背景信息提取。然后提出3通道混合注意力机制来加强局部细节信息提取,突出小型舰船与暗礁、岛屿等的差异性,提高模型细分类效果。结果在SAR图像舰船数据集HRSID(high-resolution SAR images dataset)上的实验结果表明,本文方法应用在Cascade-RCNN(cascade region convolutional neural network)、YOLOv4(you only look once v4)和BorderDet(border detection)3种检测模型上,与原模型相比,对小型舰船的检测精度分别提高了3.5%、2.6%和2.9%,总体精度达到89.9%。在SSDD(SAR ship detection dataset)数据集上的总体精度达到95.9%,优于现有方法。结论本文通过改进骨干网络,使模型能够改变卷积核形状和大小,集中获取目标信息,抑制背景信息干扰,有效降低了SAR图像近岸复杂背景下小型舰船的误检漏检情况。Objective Synthetic aperture radar(SAR)image based vessels detection is essential for marine-oriented detection and administration.Traditional constant false alarm rate(CFAR)algorithms have contributed on the targets analyses,such as reliance on hand-made features,slow speed,and susceptibility to interference from ship-like objects like roofs and containers.Convolutional neural network(CNN)based detectors have fundamentally improved detection accuracy.However,there are a large number of vessels detection results are restricted of complicated docking directions and multiple sizes in the high-resolution SAR images,so the recognition rate of the model remains low for some,especially small ships in the complex scenarios near the shore.Using the convolution kernel to extract features,the weights in the convolution kernel are multiplied with the values at the corresponding locations of the feature map.Therefore,the matching degree between the convolution kernel shape and the target shape could determine its efficiency and quality of feature extraction to a certain extent.If the shape of the convolution kernel is more similar to the target shape,the extracted feature map will contain the complete information of the target.Otherwise,the feature map will contain many background features that interfere with model classification and localization.Traditional methods are still challenged that the square convolutional kernel does not fit the shape of a ship with a long strip of random docking direction well.So,we tend to develop a backbone network based on deformable cavity convolution for that.Method Weighted fusion deformable atrous convolution(WFDAC)can somewhat adaptively change the shape and size of the convolution kernels and weight the features extracted by different convolution kernels in terms of the learned weights.In this way,the network can be made to actively learn any feature kernels are more capable of extracting features that match the target shape,thus the information-related is enhanced for the extraction of

关 键 词：舰船检测 合成孔径雷达(SAR)图像 可变形卷积 视觉注意力机制 空洞卷积 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]