期刊
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
卷 14, 期 -, 页码 1447-1457出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/JSTARS.2020.3044060
关键词
Change detection; DeepLabv3+; deep convolutional generative adversarial networks (DCGAN); deep learning; subpixel convolution
类别
资金
- Science and Technology Program of Sichuan [2017GZ0327]
- Science and Technology Program of Hebei [19255901D]
- National Defense Science and Technology Key Laboratory of Remote Sensing Information and Image Analysis Technology of China [6142A010301]
- Chinese Air-Force Equipment Pre-Research Project [10305***02]
The remote sensing image change detection method is important in various fields and this study proposed an improved deep learning network model that enhances accuracy and generalization performance through sample augmentation and network structure improvement.
Remote sensing image change detection method plays a great role in land cover research, disaster assessment, medical diagnosis, video surveillance, and other fields, so it has attracted wide attention. Based on a small sample dataset from SZTAKI Air-Change Benchmark Set, in order to solve the problem that the deep learning network needs a large number of samples, this work first uses nongenerative sample augmentation method and generative sample augmentation method based on deep convolutional generative adversarial networks, and then, constructs a remote sensing image change detection model based on an improved DeepLabv3+ network. This model can realize end-to-end training and prediction of remote sensing image change detection with subpixel convolution. Finally, Landsat 8, Google Earth, and Onera satellite change detection datasets are used to verify the generalization performance of this network. The experimental results show that the improved network accuracy is 95.1% and the generalization performance is acceptable.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据