深度卷积神经网络在迁移学习模式下的SAR目标识别

doi:10.7523/j.issn.2095-6134.2018.01.010

中国科学院大学学报 ›› 2018, Vol. 35 ›› Issue (1): 75-83.DOI: 10.7523/j.issn.2095-6134.2018.01.010

深度卷积神经网络在迁移学习模式下的SAR目标识别

李松^1,2,3, 魏中浩^1,2,3, 张冰尘^1,2, 洪文^1,2

1. 中国科学院电子学研究所, 北京 100190;
2. 微波成像技术国家级重点实验室, 北京 100190;
3. 中国科学院大学, 北京 100049

收稿日期:2017-02-10 修回日期:2017-03-28 发布日期:2018-01-15
通讯作者: 李松
基金资助:
国家自然科学基金（61571419）资助

Target recognition using the transfer learning-based deep convolutional neural networks for SAR images

LI Song^1,2,3, WEI Zhonghao^1,2,3, ZHANG Bingchen^1,2, HONG Wen^1,2

1. Institute of Electronics, Chinese Academy of Sciences, Beijing 100190, China;
2. National Key Laboratory of Microwave Imaging Technology, Beijing 100190, China;
3. University of Chinese Academy of Sciences, Beijing 100049, China

Received:2017-02-10 Revised:2017-03-28 Published:2018-01-15

摘要/Abstract

摘要： 合成孔径雷达（synthetic aperture radar，SAR）自动目标识别过程主要包括目标特征提取和分类器训练两个步骤。提出一种基于深度卷积神经网络（deep convolutional neural networks，DNNs）的SAR自动目标识别方法，使用一类优化的DNNs网络结构对SAR图像目标进行分类训练。该网络结构自动提取目标类别特征，避免人工预选取特征方法带来的不标准性。在DNNs网络模型训练过程中引入迁移学习的概念，以防止结果陷入局部最优解和加快模型参数的训练。最后使用美国运动和静止目标获取与识别MSTAR数据集进行试验，给出该方法与其他分类方法结果的对比，证明其取得较高的分类正确率。

关键词: 合成孔径雷达(SAR), 自动目标识别, 深度卷积神经网络, 迁移学习

Abstract: The automatic target recognition procedure of synthetic aperture radar (SAR) generally includes two steps, feature extraction and classifier training. Based on the development of deep convolutional neural networks, we present a new method of SAR target recognition. This method automatically learns the hierarchies of features from different targets, which means it avoids the non-normalization caused by manual feature extraction. Then the transfer learning technology is applied to avert the occurrence of locally optimal solution and accelerate the training procedure. Finally we use the moving and stationary target acquisition and recognition database to verify our method.

Key words: synthetic aperture radar(SAR), automatic target recognition(ATR), deep convolu-tional neural networks(DNNs), transfer learning

中图分类号:

TN957

李松, 魏中浩, 张冰尘, 洪文. 深度卷积神经网络在迁移学习模式下的SAR目标识别[J]. 中国科学院大学学报, 2018, 35(1): 75-83.

LI Song, WEI Zhonghao, ZHANG Bingchen, HONG Wen. Target recognition using the transfer learning-based deep convolutional neural networks for SAR images[J]. , 2018, 35(1): 75-83.

参考文献

[1] O'Sullivan J A, Devore M D, Kedia V, et al. SAR ATR performance using a conditionally Gaussian model[J]. IEEE Transactions on Aerospace & Electronic Systems, 2001, 37(1):91-108.
[2] Srinivas U, Monga V, Raj R G. SAR automatic target recognition using discriminative graphical models[J]. IEEE Transactions on Aerospace & Electronic Systems, 2014, 50(1):591-606.
[3] 张锐, 洪峻, 明峰. 基于目标CSAR回波模型的SAR自动目标识别算法[J]. 电子与信息学报, 2011, 33(1):27-32.
[4] 向卫力, 李晓辉, 周勇胜,等. 一种鲁棒的多尺度稀疏表示SAR目标识别方法[J]. 中国科学院大学学报, 2017, 34(1):99-105.
[5] Zhao Q, Principe J C. Support vector machines for SAR automatic target recognition[J]. IEEE Transactions on Aerospace & Electronic Systems, 2001, 37(2):643-654.
[6] Lecun Y, Bottou L, Bengio Y, et al. Gradient-based learning applied to document recognition[J]. Proceedings of the IEEE, 1998, 86(11):2278-2324.
[7] Chen S, Wang H, Xu F, et al. Target classification using the deep convolutional networks for SAR images[J]. IEEE Transactions on Geoscience & Remote Sensing, 2016, 54(8):4806-4817.
[8] Hinton G E, Salakhutdinov R R. Reducing the dimensionality of data with neural networks[J]. Science, 2006, 313(5786):504-507.
[9] Bengio S, Deng L, Larochelle H, et al. Special issue on learning deep architectures[J].IEEE Transactions on Pattern Analysis and Machine Intelligence, 2012, 34(1):207-207.
[10] Krizhevsky A, Sutskever I, Hinton G E. ImageNet classification with deep convolutional neural networks[C]//International Conference on Neural Information Processing Systems. Curran Associates Inc, 2012:1097-1105.
[11] Taigman Y, Yang M, Ranzato M, et al. DeepFace:closing the gap to human-level performance in face verification[C]//IEEE Conference on Computer Vision and Pattern Recognition. IEEE Computer Society, 2014:1701-1708.
[12] Sun Y, Wang X, Tang X. Deep learning face representation from predicting 10000 classes[C]//IEEE Conference on Computer Vision and Pattern Recognition. IEEE Computer Society, 2014:1891-1898.
[13] Lecun Y, Bengio Y, Hinton G. Deep learning[J]. Nature, 2015, 521(7553):436-444.
[14] Yosinski J, Clune J, Bengio Y, et al. How transferable are features in deep neural networks?[J]. Eprint Arxiv, 2014, 27:3320-3328.
[15] Li W, Duan L, Xu D, et al. Learning with augmented features for supervised and semi-supervised heterogeneous domain adaptation[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2013, 36(6):1134-1148.
[16] Duchi J, Hazan E, Singer Y. Adaptive subgradient methods for online learning and stochastic optimization[J]. Journal of Machine Learning Research, 2011, 12(7):2121-2159.
[17] Bottou L, Bousquet O. The tradeoffs of large scale learning[C]//International Conference on Neural Information Processing Systems. Curran Associates Inc, 2007:161-168.
[18] Grósz T, Gosztolya G, Tóth L. A sequence training method for deep rectifier neural networks[J]. Speech Recognition, 2014, 8773:81-88.
[19] Scherer D, Müller A, Behnke S. Evaluation of pooling operations in convolutional architectures for object recognition[C]//Artificial Neural Networks ICANN 2010, International Conference, Thessaloniki, Greece:Proceedings. DBLP, 2010:92-101.
[20] 江璐, 赵彤, 吴敏. 基于深度卷积神经网络的指纹纹型分类算法[J]. 中国科学院大学学报, 2016, 33(6):808-814.
[21] Bengio Y. Practical recommendations for gradient-based training of deep architectures[M]//Neural Networks:Tricks of the Trade. Springer Berlin Heidelberg, 2012:133-144.
[22] Pan S J, Yang Q. A survey on transfer learning[J]. IEEE Transactions on Knowledge & Data Engineering, 2009, 22(10):1345-1359.
[23] Zhao Q, Principe J C. Support vector machines for SAR automatic target recognition[J]. IEEE Transactions on Aerospace & Electronic Systems, 2001, 37(2):643-654.
[24] O'Sullivan J A, Devore M D, Kedia V, et al. SAR ATR performance using a conditionally Gaussian model[J]. IEEE Transactions on Aerospace & Electronic Systems, 2001, 37(1):91-108.
[25] Park J I, Kim K T. Modified polar mapping classifier for SAR automatic target recognition[J]. IEEE Transactions on Aerospace & Electronic Systems Aes, 2014, 50(2):1092-1107.
[26] Dong G, Kuang G. Classification on the monogenic scale space:application to target recognition in SAR image[J]. IEEE Transactions on Image Processing, 2015, 24(8):2527-2539.
[27] 张慧,肖蒙,崔宗勇.基于卷积神经网络的SAR目标多维度特征提取[J].机械制造与自动化,2017(1):111-115.
[28] 史鹤欢, 许悦雷, 马时平,等. PCA预训练的卷积神经网络目标识别算法[J]. 西安电子科技大学学报(自然科学版), 2016, 43(3):161-166.
[29] 李君宝,杨文慧,许剑清,等. 基于深度卷积网络的SAR图像目标检测识别[J]. 导航定位与授时,2017,4(1):60-66.

深度卷积神经网络在迁移学习模式下的SAR目标识别

Target recognition using the transfer learning-based deep convolutional neural networks for SAR images

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 8

编辑推荐

Metrics

本文评价

访问统计

联系我们

[1]	史达亨, 刘立刚, 周斌, 卜智勇. 跨时间迁移的多源无线信号指纹融合定位方法[J]. 中国科学院大学学报, 2021, 38(6): 817-824.
[2]	王泽隆, 徐向辉, 张雷. 基于仿真SAR图像深度迁移学习的自动目标识别[J]. 中国科学院大学学报, 2020, 37(4): 516-524.
[3]	王健飞, 张卫强, 刘加. 基于多状态跳转模型的场景独立音频事件检测方法[J]. 中国科学院大学学报, 2019, 36(2): 218-225.
[4]	肖垚, 刘畅. 基于稀疏求解的改进PCA方法在SAR目标识别中的应用[J]. 中国科学院大学学报, 2018, 35(1): 84-88.
[5]	杨洋, 王岩飞. 一种基于混合模型的合成孔径雷达自聚焦算法[J]. 中国科学院大学学报, 2016, 33(5): 656-663.
[6]	宋晓燊, 禹卫东. 条带式VideoSAR参数依赖关系的推导及应用[J]. 中国科学院大学学报, 2016, 33(1): 121-127.
[7]	关增达, 程立, 朱廷劭. 基于平移不变核的异构迁移学习[J]. 中国科学院大学学报, 2015, 32(1): 121-126.
[8]	张升康，杨汝良. 双基地合成孔径雷达扩展Chirp Scaling成像算法[J]. 中国科学院大学学报, 2008, 25(1): 101-109.