训练集对遥感图像超分辨率下自动目标识别影响的研究  

Research on the Influence of Training Set on Automatic Target Recognition Under Super-Resolution of Remote Sensing Images

作　　者：王艳[1,2] 李昂 王晟全[1,2] WANG Yan;LI Ang;WANG Shengquan(Zijin College,Nanjing University of Science and Technology,Nanjing 210023,China;School of Electronic Engineering and Optoelectronic Technology,Nanjing University of Science and Technology,Nanjing 210023,China;School of Telecommunications&Information Engineering,Nanjing University of Posts and Telecommunications,Nanjing 210003,China)

机构地区：[1]南京理工大学紫金学院,南京210023 [2]南京理工大学电子工程与光电技术学院,南京210023 [3]南京邮电大学通信学院,南京210003

出　　处：《重庆理工大学学报（自然科学）》2021年第2期136-143,共8页Journal of Chongqing University of Technology：Natural Science

基　　金：江苏省高校自然科学基金面上项目(18KJD510004);江苏省普通高校学术学位研究生科研创新计划项目(KYLX160661);江苏省高等学校大学生创新创业训练计划项目(201913654002Y)。

摘　　要：采用深度学习算法:首先通过超分辨复原生成对抗网络实现图像超分辨率,在包括测试船舶本体的多个训练集中进行训练;然后采用AlexNet网络进行分类检测;最后采用RetinaNet网络进行目标识别。实验表明:与不包含测试本体的训练集相比,包含测试本体的训练集超分辨率下的目标识别精确率最高。In this paper,deep learning algorithm is used:firstly,SISR is realized by srgan,which is trained in several training sets including testing ship ontology;secondly,alexnet network is used for classification detection;finally,retinanet network is used for target recognition.The experimental results show that,compared with the training set without test ontology,the training set with test ontology has the highest accuracy of target recognition under super-resolution.

关 键 词：图像超分辨率 遥感图像 SRGAN 目标识别 深度学习 

分 类 号：TP39[自动化与计算机技术—计算机应用技术]