[1]崔亚楠,吴建平,朱辰龙,等.改进残差网络结合迁移学习的 SAR 目标识别[J].计算机技术与发展,2022,32(05):1-6.[doi:10. 3969 / j. issn. 1673-629X. 2022. 05. 001]
　CUI Ya-nan,WU Jian-ping,ZHU Chen-long,et al.SAR Target Recognition of Improved Residual Network Combined with Transfer Learning[J].,2022,32(05):1-6.[doi:10. 3969 / j. issn. 1673-629X. 2022. 05. 001]
点击复制

改进残差网络结合迁移学习的 SAR 目标识别()
分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
32
期数:
2022年05期
页码:
1-6
栏目:
人工智能
出版日期:
2022-05-10

文章信息/Info

Title:
SAR Target Recognition of Improved Residual Network Combined with Transfer Learning
文章编号:
1673-629X(2022)05-0001-06
作者:
崔亚楠1 ; 吴建平1; 2; 3 ; 朱辰龙1 ; 闫相如1
1. 云南大学 信息学院,云南 昆明 650504;
2. 云南省电子计算中心,云南 昆明 650223;
3. 云南省高校数字媒体技术重点实验室,云南 昆明 650223
Author(s):
CUI Ya-nan1 ; WU Jian-ping1; 2; 3 ; ZHU Chen-long1 ; YAN Xiang-ru1
1. School of Information Science & Engineering,Yunnan University,Kunming 650504,China;
2. Yunnan Provincial Electronic Computing Center,Kunming 650223,China;
3. Digital Media Technology Key Laboratory of Universities and Colleges in Yunnan Province,Kunming 650223,China
关键词:
ResNet101; 迁移学习; 合成孔径雷达; 卷积注意力模块; 特征金字塔网络
Keywords:
ResNet101; transfer learning; synthetic aperture radar; convolutional block attention module; feature pyramid network
分类号:
TP753
DOI:
10. 3969 / j. issn. 1673-629X. 2022. 05. 001
摘要:
合成孔径雷达(SAR) 图像的目标识别对地面和海面目标获取具有重大意义。 实现 SAR 图像目标自动解释,提高图像目标识别的准确率成为 SAR 图像研究的热点问题。 为准确获取 SAR 图像中的目标信息,解决深度神经网络训练小样本 SAR 图像过程中细节特征丢失严重,网络易出现过拟合等问题,该研究提出一种基于 RCF( ResNet101 -CBAM-FPN)神经网络模型来提取 SAR 图像特征。 将 ResNet101 作为主干网络模型用于特征提取,在主干网络模型中加入卷积注意力模块引导神经网络有针对性地提取 SAR 图像关键特征信息。 然后结合特征金字塔网络,实现神经网络高层特征与底层特征融合,丰富特征信息。 最后融合迁移学习思想,通过数据相对充足的仿真 SAR 图像对 RCF 网络模型进行预训练。 将预训练获取的模型参数迁移至目标网络,作为目标网络的初始化参数,并使用目标网络对 SAR 图像进行迭代训练。 实验结果表明,该方法能有效提升小样本数据 SAR 图像的识别精度,在 MSTAR 数据集上达到 99. 60% 的识别率。
Abstract:
The target recognition of synthetic aperture radar ( SAR) images is of great significance to the acquisition of military targets on the ground and sea. Realizing the automatic interpretation of SAR image targets and improving the accuracy of image target recognition have become a hot? issue in SAR image research. To accurately obtain the target information in the SAR image,and solve the problem of serious loss of detailed features in the process of deep neural network training small sample SAR images,and the network is prone to over-fitting,we propose a neural network model based on RCF ( ResNet101-CBAM-FPN) to extract SAR image features. ResNet101 is used as the backbone network model for feature extraction, and the convolutional attention module is added to the backbone network model to guide the neural network to be targeted for extraction of key feature information of SAR images. Then,combined the feature pyramid network,the fusion of high-level features and low-level features of neural network is realized to enrich feature information. Finally,the ideas of transfer learning is fused. The RCF network model is pre-trained by simulating SAR images with relatively sufficient data. The model parameters obtained by pre-training are transferred to the target network as the initialization parameters of the target network,and the target network is used to iteratively train the SAR image. The experiment shows that the proposed method can effectively improve the recognition accuracy of SAR images with small sample data,and achieve a high recognition rate of 99. 60% on the MSTAR data set.

相似文献/References:

[1]李 勇,刘战东,张海军.跨项目软件缺陷预测方法研究综述[J].计算机技术与发展,2020,30(03):98.[doi:10. 3969 / j. issn. 1673-629X. 2020. 03. 019]
　LI Yong,LIU Zhan-dong,ZHANG Hai-jun.Review on Cross-project Software Defects Prediction Methods[J].,2020,30(05):98.[doi:10. 3969 / j. issn. 1673-629X. 2020. 03. 019]
[2]武苏雯,赵慧杰,刘 鑫,等.基于迁移学习的图像分类在诗词中的应用研究[J].计算机技术与发展,2021,31(07):215.[doi:10. 3969 / j. issn. 1673-629X. 2021. 07. 036]
　WU Su-wen,ZHAO Hui-jie,LIU Xin,et al.Research on Application of Image Classification Based onTransfer Learning in Poetry[J].,2021,31(05):215.[doi:10. 3969 / j. issn. 1673-629X. 2021. 07. 036]
[3]娄丰鹏,吴迪,荆晓远,等.增加度量元的迁移学习跨项目软件缺陷预测[J].计算机技术与发展,2018,28(07):103.[doi:10.3969/ j. issn.1673-629X.2018.07.022]
　LOU Feng-peng,WU Di,JING Xiao-yuan,et al.Cross-project Software Defect Prediction Based on Transfer Learning with Metrics[J].,2018,28(05):103.[doi:10.3969/ j. issn.1673-629X.2018.07.022]
[4]刘宇廷,倪颖杰.融合知识迁移学习的微博社团检测模型构建[J].计算机技术与发展,2018,28(09):11.[doi:10.3969/j.issn.1673-629X.2018.09.003]
　LIU Yu-ting,NI Ying-jie.Construction of Weibo Community Detection Model with Knowledge Transfer Learning[J].,2018,28(05):11.[doi:10.3969/j.issn.1673-629X.2018.09.003]
[5]王泽泓,刘厚泉.基于迁移学习与自适应特征融合的建筑物识别[J].计算机技术与发展,2019,29(12):40.[doi:10. 3969 / j. issn. 1673-629X. 2019. 12. 007]
　WANG Ze-hong,LIU Hou-quan.Building Recognition Based on Transfer Learning and Adaptive Feature Fusion[J].,2019,29(05):40.[doi:10. 3969 / j. issn. 1673-629X. 2019. 12. 007]
[6]易 未,郑沫利,赵艳轲,等.基于小样本 SVR 的迁移学习及其应用[J].计算机技术与发展,2020,30(02):47.[doi:10. 3969 / j. issn. 1673-629X. 2020. 02. 010]
　YI Wei,ZHENG Mo-li,ZHAO Yan-ke,et al.Transfer Learning Based on Support Vector Regression Model for Small Sample Data and Its Applications[J].,2020,30(05):47.[doi:10. 3969 / j. issn. 1673-629X. 2020. 02. 010]
[7]王新美,丁爱玲,雷梦宁,等.基于 CNN 和 SVM 融合的交通标志识别[J].计算机技术与发展,2020,30(06):7.[doi:10. 3969 / j. issn. 1673-629X. 2020. 06. 002]
　WANG Xin-mei,DING Ai-ling,LEI Meng-ning,et al.Traffic Sign Recognition Based on Combination of CNN and SVM[J].,2020,30(05):7.[doi:10. 3969 / j. issn. 1673-629X. 2020. 06. 002]
[8]龚 安,井晓萌.多卷积神经网络模型融合的农作物病害图像识别[J].计算机技术与发展,2020,30(08):134.[doi:10. 3969 / j. issn. 1673-629X. 2020. 08. 023]
　GONG An,JING Xiao-meng.Image Recognition of Crop Diseases Based on Multi-convolution Neural Network Model Ensemble[J].,2020,30(05):134.[doi:10. 3969 / j. issn. 1673-629X. 2020. 08. 023]
[9]龚 安,郭文婷.基于卷积神经网络的皮肤癌识别方法[J].计算机技术与发展,2020,30(10):167.[doi:10. 3969 / j. issn. 1673-629X. 2020. 10. 030]
　GONG An,GUO Wen-ting.Skin Cancer Image Classification Method Based on Convolutional Neural Network[J].,2020,30(05):167.[doi:10. 3969 / j. issn. 1673-629X. 2020. 10. 030]
[10]张泽宇,郭 斌,张太红*.基于 DCNN 的马匹图像分割算法研究[J].计算机技术与发展,2020,30(10):210.[doi:10. 3969 / j. issn. 1673-629X. 2020. 10. 037]
　ZHANG Ze-yu,GUO Bin,ZHANG Tai-hong.Research on Horse Image Segmentation Algorithm Based on DCNN[J].,2020,30(05):210.[doi:10. 3969 / j. issn. 1673-629X. 2020. 10. 037]

更新日期/Last Update: 2022-05-10