此数据集是一个关于青藏高原的湖泊水体可见光谱谷歌地球遥感数据集,范围包含整个青藏高原,且是由RGB图像组成。此数据集共包含有6774张图像,大小为256* 256, DPI为96,深度为24。在此数据集中,只有湖泊(不包括河流、水库、池塘等)被注释,且所有的这些注释都是使用labelme进行标记。
采集时间 | 2020/08/01 - 2020/08/31 |
---|---|
采集地点 | 青藏高原 |
海拔 | 3000.0m - 5000.0m |
数据量 | 520.5 MiB |
数据空间分辨率(/米) | 17m |
坐标系 |
谷歌地球
为保证模型对样本的学习能力,我们将图像分割成不重叠的块,并使用数据平衡策略,若图像中负样本占比太高,则丢弃图像。
256*256
# | 编号 | 名称 | 类型 |
1 | E01Z7902 | 国家冰川冻土沙漠科学数据中心 | 其他 |
2 | XXH-13514-0209 | 中国科学院冰川冻土沙漠数据中心能力建设 | 其他 |
# | 标题 | 文件大小 |
---|---|---|
1 | train_img.zip | 513.5 MiB |
2 | train_label.zip | 7.0 MiB |
3 | 数据说明.txt | 1.5 KiB |
# | 时间 | 姓名 | 用途 |
---|---|---|---|
1 | 2024/04/22 03:16 | 周*龙 |
作为实验数据,复现论文 已检验模型性能,
|
2 | 2024/03/26 19:02 | 孙*宇 |
用于研究联邦学习下的二分类水体提取语义分割任务
|
3 | 2024/03/20 07:30 | 赵*迪 |
Paper title:基于深度学习的SAR影像青藏高原多年冻土区热融湖提取
Paper abstract:训练集
Paper type:本科毕业论文
Tutor:刘修国
|
4 | 2024/02/23 00:02 | w* |
毕设毕设毕设毕设毕设毕设毕设毕设毕设毕设毕设毕设毕设毕设毕设毕设毕设毕设毕设
|
5 | 2024/02/18 22:56 | 高*鹏 |
科研需要,需要做关于水体提取的研究,我们认为这个数据集是非常合适的
|
6 | 2024/02/07 00:53 | 刘* |
Paper title:A Novel Deep Learning Network Model for Extracting Lake Water Bodies from Remote Sensing Images
Paper abstract:Extraction of lake water bodies from remote sensing images provides reliable data support for water resource management, environmental protection, natural disaster early warning, and scien-tific research, and helps to promote sustainable development, protect the ecological environment and human health. With reference to the classical encoding-decoding semantic segmentation net-work, we propose the network model R50A3-LWBENet for lake water body extraction from re-mote sensing images based on ResNet50 and three attention mechanisms. R50A3-LWBENet model uses ResNet50 for feature extraction, also known as encoding, and squeeze and excitation (SE) block is added to the residual module, which highlights the deeper features of the water body part of the feature map during the down-sampling process, and also takes into account the importance of the feature map channels, which can better capture the multiscale relationship between pixels. After the feature extraction is completed, the convolutional block attention module (CBAM) is added to give the model a global adaptive perception capability and pay more attention to the water body part of the image. The feature map is up-sampled using bilinear interpolation, and the features at different levels are fused, a process also known as decoding, to finalize the extraction of the lake water body. Compared with U-Net, AU-Net, RU-Net, ARU-Net, SER34AUNet, and MU-Net, the R50A3-LWBENet model has the fastest convergence speed and the highest MIoU ac-curacy with a value of 97.6%, which is able to better combine global and local information, refine the edge contours of the lake’s water body, and have stronger feature extraction capability and segmentation performance.
Paper type:期刊论文
Tutor:刘江平
|
7 | 2024/01/20 03:25 | 郭*佳 |
完成学校布置的论文做实验,需要相关的数据集
|
8 | 2024/01/05 23:46 | 张*晨 |
论文题目:面向遥感水域分割的深度学习困难感知模型研究
数据在研究中的作用:用于模型训练与测试,并进行结果分析
论文类型:硕士论文
导师姓名:王蓉芳
|
9 | 2024/01/03 17:56 | 李* |
用于高原变化检测模型测试,本人打算尝试公开数据集的效果
|
10 | 2023/12/13 04:01 | 施* |
论文题目:尚未确定
数据在研究中的作用:期末作业要引用一篇专业的论文进行Paper reflection,需要对引用的论文中的数据进行分析和数据可视化
论文类型:paper reflection
导师姓名:王术
|
# | 类别 | 标题 | 作者 | 年份 |
---|---|---|---|---|
1 | 论文 | MSLWENet: A Novel Deep Learning Network for LakeWater BodyExtraction of Google Remote Sensing Images | 王兆滨、高雄 | 2020 |
© 中国科学院西北生态环境资源研究院 2005- 备案号:陇ICP备2021001824号-21
兰州市东岗西路320号 730000 电话:0931-4967592, 0931-4967596