首页 > 论文 > 光学学报 > 37卷 > 10期(pp:1015002--1)

特征融合的卷积神经网络多波段舰船目标识别

Convolutional Neural Network Based Multi-Band Ship Target Recognition with Feature Fusion

  • 摘要
  • 论文信息
  • 参考文献
  • 被引情况
  • PDF全文
分享:

摘要

针对海面背景舰船目标单一波段图像识别率低的问题, 提出了一种基于卷积神经网络(CNN)的融合识别方法。该方法提取可见光、中波红外和长波红外3个波段舰船目标特征进行融合识别。模型主要分为3个步骤:通过设计的6层CNN, 同时对三波段图像进行特征提取; 利用基于互信息的特征选择方法对串联的三波段特征向量按照重要性进行排序, 并按照图像清晰度评价指标选取固定长度的特征向量作为目标识别依据; 通过额外的2个全连接层和输出层进行回归训练。采用自建的三波段舰船图像数据库进行模型的训练和测试, 共包含6类目标, 5000余张图像。实验结果表明, 本文方法识别率达到84.5%, 与单波段识别方法相比有明显提升。

Abstract

In order to improve the recognition rate in single-band images of ship targets with complex background, we propose a new fusion recognition method based on convolutional neural networks (CNN). This method extracts the ship target features of images in three wave bands, which are visible light, medium-wave infrared and long-wave infrared images. The model is divided into three steps. Firstly, a 6-layer CNN model is designed to extract the image features of three bands simultaneously. Secondly, a feature selection method based on mutual information is used for sorting the concatenated features according to the importance, and then the feature vectors of fixed dimension can be chosen depending on the indicator of image clarity evaluation. The dimension-reduced feature vector is regarded as the basis of target recognition. Finally, a 2-layer fully connected networks and an output layer are designed for training and regression. We build a triple-band ship target dataset for our experimental verification, which contains 6 categories of targets and more than 5000 images. The experimental results show that the recognition rate of the proposed method can reach 84.5%, which is improved significantly compared to that of the single-band recognition method.

Newport宣传-MKS新实验室计划
补充资料

中图分类号:TP391

DOI:10.3788/AOS201737.1015002

所属栏目:机器视觉

基金项目:国家自然科学基金(61303192)

收稿日期:2017-04-10

修改稿日期:2017-05-15

网络出版日期:--

作者单位    点击查看

刘 峰:海军航空工程学院控制工程系, 山东 烟台 264001
沈同圣:中国国防科技信息中心, 北京 100142
马新星:海军航空工程学院控制工程系, 山东 烟台 264001

联系人作者:刘峰(liufeng_cv@126.com)

备注:刘 峰(1988-), 男, 博士研究生, 主要从事计算机视觉、图像处理、目标检测等方面的研究。

【1】Zhang Difei, Zhang Jinsuo, Yao Keming, et al. Infrared ship-target recognition based on SVM classification[J]. Infrared and Laser Engineering, 2016, 45(1): 0104004.
张迪飞, 张金锁, 姚克明, 等. 基于SVM分类的红外舰船目标识别[J]. 红外与激光工程, 2016, 45(1): 0104004.

【2】Ramanan D, Zhu X. Face detection, pose estimation, and landmark localization in the wild[C]. IEEE Conference on Computer Vision and Pattern Recognition, 2012: 2879-2886.

【3】Wojek C, Dollár P, Schiele B, et al. Pedestrian detection: an evaluation of the state of the art[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2012, 34(4): 743-761.

【4】Feineigle P A, Morris D D, Snyder F D. Ship recognition using optical imagery for harbor surveillance[C]. Proceedings of AUVSI Unmanned Systems North America Conference, 2007: 249-263.

【5】Sánchez J, Perronnin F, Mensink T, et al. Image classification with the fisher vector: theory and practice[J]. International Journal of Computer Vision, 2013, 105(3): 222-245.

【6】Smeelen M A, Schwering P B W, Toet A, et al. Semi-hidden target recognition in gated viewer images fused with thermal IR images[J]. Information Fusion, 2014, 18: 131-147.

【7】Zhou Feiyan, Jin Linpeng, Dong Jun. Review of convolutional neural network[J]. Chinese Journal of Computers, 2017, 40(6): 1229-1251.
周飞燕, 金林鹏, 董军. 卷积神经网络研究综述[J]. 计算机学报, 2017, 40(6): 1229-1251.

【8】Krizhevsky A, Sutskever I, Hinton G E. ImageNet classification with deep convolutional neural networks[C]. International Conference on Neural Information Processing Systems, 2012: 1097-1105.

【9】Liu Dawei, Han Ling, Han Xiaoyong. High spatial resolution remote sensing image classification based on deep learning[J]. Acta Optica Sinica, 2016, 36(4): 0428001.
刘大伟, 韩玲, 韩晓勇. 基于深度学习的高分辨率遥感影像分类研究[J]. 光学学报, 2016, 36(4): 0428001.

【10】He K, Zhang X, Ren S, et al. Deep residual learning for image recognition[C]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2016: 770-778.

【11】Ren S, He K, Girshick R, et al. Faster R-CNN: towards real-time object detection with region proposal networks[C]. Advances in Neural Information Processing Systems, 2015: 91-99.

【12】Redmon J, Divvala S, Girshick R, et al. You only look once: unified, real-time object detection[J/OL]. (2016-05-09)[2017-01-05] https:∥arxiv.org/abs/1506.02640.

【13】Kuen J, Wang Z, Wang G. Recurrent attentional networks for saliency detection[C]. IEEE Conference on Computer Vision and Pattern Recognition, 2016: 3668-3677.

【14】Bousetouane F, Morris B. Fast CNN surveillance pipeline for fine-grained vessel classification and detection in maritime scenarios[C]. IEEE International Conference on Advanced Video and Signal Based Surveillance, 2016: 242-248.

【15】Zhang M M, Choi J, Daniilidis K, et al. VAIS: a dataset for recognizing maritime imagery in the visible and infrared spectrums[C]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops, 2015: 10-16.

【16】Nair V, Hinton G E. Rectified linear units improve restricted Boltzmann machines[C]. Proceedings of the 27th International Conference on Machine Learning, 2010: 807-814.

【17】Zhang Y, Wu J, Cai J. Compact representation for image classification: to choose or to compress [C]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2014: 907-914.

【18】Ye P, Kumar J, Doermann D. Beyond human opinion scores: blind image quality assessment based on synthetic scores[C]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2014: 4241-4248.

【19】Xue W, Zhang L, Mou X, et al. Gradient magnitude similarity deviation: a highly efficient perceptual image quality index[J]. IEEE Transactions on Image Processing, 2014, 23(2): 684-695.

【20】Sheikh H R, Bovik A C, de Veciana G. An information fidelity criterion for image quality assessment using natural scene statistics[J]. IEEE Transactions on Image Processing, 2005, 14(12): 2117-2128.

【21】Zhang L, Zhang L, Mou X, et al. FSIM: a feature similarity index for image quality assessment[J]. IEEE Transactions on Image Processing, 2011, 20(8): 2378-2386.

【22】Wang Z, Bovik A C, Sheikh H R, et al. Image quality assessment: from error visibility to structural similarity[J]. IEEE Transactions on Image Processing, 2004, 13(4): 600-612.

引用该论文

Liu Feng,Shen Tongsheng,Ma Xinxing. Convolutional Neural Network Based Multi-Band Ship Target Recognition with Feature Fusion[J]. Acta Optica Sinica, 2017, 37(10): 1015002

刘 峰,沈同圣,马新星. 特征融合的卷积神经网络多波段舰船目标识别[J]. 光学学报, 2017, 37(10): 1015002

被引情况

【1】徐海洋,孔军,蒋敏,昝宝锋. 基于时空方向主成分直方图的人体行为识别. 激光与光电子学进展, 2018, 55(6): 61009--1

【2】王文秀,傅雨田,董峰,李锋. 基于深度卷积神经网络的红外船只目标检测方法. 光学学报, 2018, 38(7): 712006--1

【3】刘坤,苏彤,王典. 基于模糊不变卷积神经网络的遥感飞机识别. 激光与光电子学进展, 2018, 55(8): 82001--1

【4】曹宇剑,徐国明,史国川. 基于旋转不变Faster R-CNN的低空装甲目标检测. 激光与光电子学进展, 2018, 55(10): 101501--1

【5】华夏,王新晴,王东,马昭烨,邵发明. 基于改进SSD的交通大场景多目标检测. 光学学报, 2018, 38(12): 1215003--1

【6】寇添,周中良,刘宏强,杨远志,阮铖巍. 空中点目标机动模式的双色比特征空间特性及辨识. 光学学报, 2018, 38(12): 1204001--1

【7】陶志勇,王浩童,王藜谚. 一种主动鉴别未知类别指静脉的识别系统. 激光与光电子学进展, 2018, 55(12): 121002--1

【8】张肃,战俊彤,付强,段锦,姜会林. 基于多小波融合的偏振探测去雾技术. 激光与光电子学进展, 2018, 55(12): 122602--1

【9】王书宇,陶声祥,杨钒,艾磊. 基于卷积神经网络的激光距离选通式成像目标识别. 激光与光电子学进展, 2019, 56(2): 21001--1

【10】郑婷月,唐晨,雷振坤. 基于全卷积神经网络的多尺度视网膜血管分割. 光学学报, 2019, 39(2): 211002--1

【11】任之俊,蔺素珍,李大威,王丽芳,左健宏. 基于改进特征金字塔的Mask R-CNN目标检测方法. 激光与光电子学进展, 2019, 56(4): 41502--1

【12】陈海永,郄丽忠,刘坤. 基于区域辐射一致性的移动阴影检测. 光学学报, 2019, 39(3): 315003--1

【13】马永杰,马芸婷,陈佳辉. 结合卷积神经网络多层特征和支持向量机的车辆识别. 激光与光电子学进展, 2019, 56(14): 141001--1

【14】李灏天,陈晓冬,徐怀远,许鸿雁,汪毅,蔡怀宇. 基于Bilateral-Frangi滤波的桥梁裂缝检测算法. 激光与光电子学进展, 2019, 56(18): 181401--1

【15】朱天佑,黄凌锋,董峰,龚惠兴. 基于轻量级残差网络的红外遥感船只检测. 光学学报, 2020, 40(1): 111018--1

【16】张晓龙,刘剑飞,郝禄国. 基于特征密集计算与融合算法的教师课堂行为分析. 激光与光电子学进展, 2019, 56(16): 161503--1

【17】李卓容,王凯旋,何欣龙,糜忠良,唐云祁. 基于卷积神经网络的足跟着地事件检测算法. 激光与光电子学进展, 2019, 56(21): 211503--1

您的浏览器不支持PDF插件,请使用最新的(Chrome/Fire Fox等)浏览器.或者您还可以点击此处下载该论文PDF