| 类别非均衡遥感图像语义分割的全卷积网络方法 |
| |
| 引用本文： | 吴止锾,高永明,李磊,薛俊诗.类别非均衡遥感图像语义分割的全卷积网络方法[J].光学学报,2019,39(4):393-404. |
| |
| 作者姓名： | 吴止锾 高永明 李磊 薛俊诗 |
| |
| 作者单位： | 航天工程大学研究生院,北京101416;63883部队,河南洛阳471000;航天工程大学航天信息学院,北京,101416;航天工程大学电子与光学工程系,北京,101416;航天工程大学研究生院,北京,101416 |
| |
| 摘 要： | 基于U-Net模型,提出了一个全卷积网络(FCN)模型,用于高分辨率遥感图像语义分割,其中数据预处理采用了数据标准化和数据增强,模型训练过程采用Adam优化器,模型性能评估采用平均Jaccard指数。为提高小类预测的准确率,模型中采用了加权交叉熵损失函数和自适应阈值方法。在DSTL数据集上进行了实验,结果表明所提方法将预测结果的平均Jaccard指数从0.611提升到0.636,可实现对高分辨率遥感图像端到端的精确分类。
|
| 关 键 词： | 图像处理 遥感图像 语义分割 类别非均衡 全卷积网络 深度学习 |
Fully Convolutional Network Method of Semantic Segmentation of Class Imbalance Remote Sensing Images |
| |
| Wu Zhihuan,Gao Yongming,Li Lei,Xue Junshi.Fully Convolutional Network Method of Semantic Segmentation of Class Imbalance Remote Sensing Images[J].Acta Optica Sinica,2019,39(4):393-404. |
| |
| Authors: | Wu Zhihuan Gao Yongming Li Lei Xue Junshi |
| |
| Institution: | (Graduate School, Space Engineering University, Beijing 101416, China;63883 Troops, Luoyang. Henan 471000, China;School of Space Information , Space Engineering University, Beijing 101416, China;Department of Electronic and Optical Engineering, Space Engineering University, Beijing 101416, China) |
| |
| Abstract: | Wu Zhihuan;Gao Yongming;Li Lei;Xue Junshi(Graduate School, Space Engineering University, Beijing 101416, China;63883 Troops, Luoyang. Henan 471000, China;School of Space Information , Space Engineering University, Beijing 101416, China;Department of Electronic and Optical Engineering, Space Engineering University, Beijing 101416, China) |
| |
| Keywords: | image processing:remote sensing images semantic segmentation class imbalance:fully convolutional network deep learning |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
