您的位置:首页  > 论文页面

基于深度学习的遥感图像语义分割预测增强技术

发表时间:2021-03-31  浏览量:299  下载量:57
全部作者: 朱梁辉,罗小波
作者单位: 重庆邮电大学计算机科学与技术学院
摘 要: 深度学习是实现遥感图像自动解译的重要途径,但遥感图像尺度往往较大,需要剪裁成小尺度图像进行训练预测,而不同的预测策略将影响最终的预测效果。本文论述了遥感图像语义分割的4种预测策略,并通过在国际摄影测量与遥感学会(International Society for Photogrammetry and Remote Sensing,ISPRS)波茨坦数据集上训练全卷积网络(fully convolutional network,FCN)、U-Net、SegNet和DenseASPP模型,对这些预测策略进行了评估。研究结果显示,使用预测增强技术相比不使用预测增强大约能提高2%~3%的分割精度,但随之而来的是需要消耗更多的计算资源。
关 键 词: 摄影测量与遥感技术;语义分割;遥感图像;测试时增强;预测优化
Title: Remote sensing image semantic segmentation prediction augmentation based on deep learning
Author: ZHU Lianghui, LUO Xiaobo
Organization: College of Computer Science and Technology, Chongqing University of Posts and Telecommunications
Abstract: Deep learning is an important way to realize automatic interpretation of remote sensing images. However, the scale of remote sensing images is often large, such large-scale images need to be cropped into small-scale images for training and prediction, and different prediction strategies will affect the final prediction effect. In this paper, four prediction strategies for remote sensing image semantic segmentation are discussed, and fully convolutional network (FCN), U-Net, SegNet and DenseASPP models are trained on the International Society for Photogrammetry and Remote Sensing (ISPRS) Potsdam dataset to evaluate these prediction strategies. The research result shows that the use of predictive enhancement technology can improve the segmentation accuracy by about 2%-3% compared with not using predictive enhancement, but it will consume more computing resources.
Key words: photogrammetry and remote sensing technology; semantic segmentation; remote sensing image; test-time augmentation; prediction optimization
发表期数: 2021年3月第1期
引用格式: 朱梁辉,罗小波. 基于深度学习的遥感图像语义分割预测增强技术[J]. 中国科技论文在线精品论文,2021,14(1):106-113.
 
6 评论数 0
暂无评论
友情链接