首页 | 本学科首页   官方微博 | 高级检索  
     检索      

水下高分辨率声图中小目标的深度网络分类方法
引用本文:朱可卿,田杰,黄海宁.水下高分辨率声图中小目标的深度网络分类方法[J].声学学报,2019,44(4):595-603.
作者姓名:朱可卿  田杰  黄海宁
作者单位:1. 中国科学院声学研究所 北京 100190;
基金项目:“问海计划”项目(SQ2017WHZZB0701-3-2)资助
摘    要:针对声成像数据缺少条件下的水下沉底小目标分类问题,提出一种深度网络分类算法。首先,采用高斯混合模型对声影区统计特性进行建模并提取声图阴影,在此基础上构建仿真数据集和真实数据集。将仿真数据集输入卷积神经网络进行训练,保留其特征提取部分,用于对真实数据集进行特征提取.重建网络分类部分并采用真实数据集的特征向量进行训练。结果表明,所提出的方法分类正确率可达88.24%,与6种对照方法相比平均分类正确率分别提升8.67%,20.47%,19.78%,11.59%,9.01%,11.58%。验证了所提出方法在小样本条件下具有较好对水下沉底小目标的分类能力。其学习曲线收敛到96.25%,仅比验证曲线高5.14%,说明在一定程度上缓解了过拟合问题。将改进的卷积神经网络应用于融合分类器,通过与逻辑回归分类器、支持向量机对目标进行分类并融合决策,正确率为93.33%,可进一步提高算法的正确率和稳定性. 

关 键 词:水下高分辨率成像声呐    水下小目标分类    深度网络声图分类
收稿时间:2018-09-27

Underwater objects classification method in high-resolution sonar images using deep neural network
Institution:1. Institute of Acoustics, Chinese Academy of Sciences Beijing 100190;2. Key Laboratory of Science and Technology on Advanced Underwater Acoustic Signal Processing, Chinese Academy of Sciences Beijing 100190;3. University of Chinese Academy of Sciences Beijing 100049
Abstract:To solve the problem of underwater proud object classification under small sample size situation where sonar image data is lacked,a classification method using deep neural network is proposed.Firstly,statistical characteristics of acoustic shadow regions are modeled using Gaussian mixture model and acoustic shadows are extracted.Trial and simulated dataset are constructed on this basis.Then,simulated dataset is input into convolutional neural network for training,and the feature extraction part is retained,which is used to extract feature of trial dataset.The classification part is reconstructed and trained by feature vectors of trial dataset.The experimental results show that the average classification accuracy of the proposed method is 88.24%,which is 8.67%,20.47%,19.78%,11.59%,9.01%,11.58%higher than that of other six contrast methods respectively.It verifies that the proposed method achieves better performance on underwater proud object classification problem.The learning curve converges to an accuracy of 96.25%,which is only 5.14% higher than the validation curve,indicating that the over-fitting problem is alleviated to some extent.Improved convolutional neural network is applied in a fusion classifier,which also combines the output of Logistic Regression classifier,support vector machine,and finally obtains a fusion result.The classification accuracy is up to 93.33%,indicating that fusion classifier improves the robustness and classification performance of the algorithm further. 
Keywords:
本文献已被 CNKI 等数据库收录!
点击此处可从《声学学报》浏览原始摘要信息
点击此处可从《声学学报》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号