北京航空航天大学学报 ›› 2017, Vol. 43 ›› Issue (9): 1841-1848.doi: 10.13700/j.bh.1001-5965.2016.0755

• 论文 • 上一篇    下一篇

基于卷积神经网络的遥感图像舰船目标检测

黄洁1,2, 姜志国1,2, 张浩鹏1,2, 姚远1,2   

  1. 1. 北京航空航天大学 宇航学院, 北京 100083;
    2. 北京航空航天大学 数字媒体北京市重点实验室, 北京 100083
  • 收稿日期:2016-09-26 出版日期:2017-09-20 发布日期:2017-04-07
  • 通讯作者: E-mail:jiangzg@buaa.edu.cn E-mail:jiangzg@buaa.edu.cn
  • 作者简介:黄洁,女,硕士研究生;主要研究方向:图像目标检测;姜志国,男,博士,教授,博士生导师;主要研究方向:遥感图像处理、医学图像处理;张浩鹏,男,博士,讲师;主要研究方向:目标检测识别、三维重建及姿态测量;姚远,男,博士研究生;主要研究方向:遥感图像目标检测与识别
  • 基金资助:
    国家重点研发计划(2016YFB0501300,2016YFB0501302);国家自然科学基金(61501009,61371134,61071137);航天科技创新基金

Ship object detection in remote sensing images using convolutional neural networks

HUANG Jie1,2, JIANG Zhiguo1,2, ZHANG Haopeng1,2, YAO Yuan1,2   

  1. 1. School of Astronautics, Beijing University of Aeronautics and Astronautics, Beijing 100083, China;
    2. Beijing Key Laboratory of Digital Media, Beijing University of Aeronautics and Astronautics, Beijing 100083, China
  • Received:2016-09-26 Online:2017-09-20 Published:2017-04-07
  • Supported by:
    National Key Research and Development Program of China (2016YFB0501300, 2016YFB0501302); National Natural Science Foundation of China (61501009, 61371134, 61071137); Aerospace Science and Technology Innovation Fund of CASC

摘要: 针对遥感图像背景复杂、受环境因素影响大的问题,提出一种将卷积神经网络(CNN)与支持向量机(SVM)相结合的舰船目标检测方法,利用卷积神经网络可自主提取图像特征并进行学习的优点,避免了复杂的特征选择和提取过程,在复杂海况背景图像的处理中体现出较优的性能;同时,由于军舰样本获取难度大,应用迁移学习的概念,利用大量民船样本辅助军舰目标的检测,取得较好的效果。通过参数调整与实验验证,此方法在自行建立的测试集上检测率达到90.59%,对光照、环境等外界因素具有一定程度的鲁棒性。

关键词: 卷积神经网络(CNN), 支持向量机(SVM), 舰船检测, 特征提取, 迁移学习

Abstract: Object detection in remote sensing images is mostly suffered from complex background and multiple interferences of environment. In this paper, a new method of ship detection is proposed, which combines convolutional neural networks (CNN) and support vector machine (SVM) to complete the ship detection task. Convolutional layers were adopted for feature extraction, taking advantages of independent feature extraction of CNNs and avoiding the process of complicated feature selection and extraction, which leads to better detection performance in complex background images. Meanwhile, since the samples of warship are difficult to acquire, samples of civil ship were employed as assistant samples for warship detection based on transfer learning theory. And this transfer learning method is proved to be effective by the experimental results, which performs better than the model trained only with warship samples. According to the parameter tuning and experimental validation, this method achieves a precision of 90.59% on testing dataset established by ourselves. In conclusion, this method possesses feasibility and robustness under different conditions of illumination and environment.

Key words: convolutional neural networks (CNN), support vector machine (SVM), ship detection, feature extraction, transfer learning

中图分类号: 


版权所有 © 《北京航空航天大学学报》编辑部
通讯地址:北京市海淀区学院路37号 北京航空航天大学学报编辑部 邮编:100191 E-mail:jbuaa@buaa.edu.cn
本系统由北京玛格泰克科技发展有限公司设计开发