TGRS-HRRSD-Dataset-master
所属分类：其他
开发工具：Python
文件大小：19131KB
下载次数：5
上传日期：2019-10-24 10:25:58
上 传 者：
dreamkkk
说明： 深度学习目标识别样本，已经做好注记，注记为xml格式可以直接在yolov3等网络进行训练
(Deep learning target recognition samples have been annotated. The XML format can be directly trained in yolov3 and other networks.)
文件列表:
OPT2017 (0, 2019-09-10)
OPT2017\Annotations (0, 2019-09-10)
OPT2017\Annotations\00001.xml (666, 2019-09-10)
OPT2017\Annotations\00002.xml (1807, 2019-09-10)
OPT2017\Annotations\00003.xml (667, 2019-09-10)
OPT2017\Annotations\00004.xml (2034, 2019-09-10)
OPT2017\Annotations\00005.xml (1576, 2019-09-10)
OPT2017\Annotations\00006.xml (1345, 2019-09-10)
OPT2017\Annotations\00007.xml (2268, 2019-09-10)
OPT2017\Annotations\00008.xml (1122, 2019-09-10)
OPT2017\Annotations\00009.xml (4088, 2019-09-10)
OPT2017\Annotations\00010.xml (1348, 2019-09-10)
OPT2017\Annotations\00011.xml (1576, 2019-09-10)
OPT2017\Annotations\00012.xml (2492, 2019-09-10)
OPT2017\Annotations\00013.xml (2474, 2019-09-10)
OPT2017\Annotations\00014.xml (666, 2019-09-10)
OPT2017\Annotations\00015.xml (2926, 2019-09-10)
OPT2017\Annotations\00016.xml (1352, 2019-09-10)
OPT2017\Annotations\00017.xml (897, 2019-09-10)
OPT2017\Annotations\00018.xml (1350, 2019-09-10)
OPT2017\Annotations\00019.xml (1810, 2019-09-10)
OPT2017\Annotations\00020.xml (1343, 2019-09-10)
OPT2017\Annotations\00021.xml (896, 2019-09-10)
OPT2017\Annotations\00022.xml (1576, 2019-09-10)
OPT2017\Annotations\00023.xml (1569, 2019-09-10)
OPT2017\Annotations\00024.xml (670, 2019-09-10)
OPT2017\Annotations\00025.xml (1569, 2019-09-10)
OPT2017\Annotations\00026.xml (1117, 2019-09-10)
OPT2017\Annotations\00027.xml (1343, 2019-09-10)
OPT2017\Annotations\00028.xml (1793, 2019-09-10)
OPT2017\Annotations\00029.xml (668, 2019-09-10)
OPT2017\Annotations\00030.xml (1118, 2019-09-10)
OPT2017\Annotations\00031.xml (2498, 2019-09-10)
OPT2017\Annotations\00032.xml (1818, 2019-09-10)
OPT2017\Annotations\00033.xml (670, 2019-09-10)
OPT2017\Annotations\00034.xml (894, 2019-09-10)
OPT2017\Annotations\00035.xml (1118, 2019-09-10)
OPT2017\Annotations\00036.xml (890, 2019-09-10)
OPT2017\Annotations\00037.xml (1341, 2019-09-10)
... ...
TGRS-HRRSD-Dataset: *High Resolution Remote Sensing Detection* (HRRSD)
=====================
- HRRSD contains **21,761 images** acquired from Google Earth and Baidu Map with the spatial resolution from 0.15-m to 1.2-m.
- There are **55,740 object instances** in HRRSD.
- HRRSD contains **13 categories** of RSI objects.
Moreover, this dataset is divided as several subsets, image numbers in each subset are **5401 for ‘train’, 5417 for ‘val’, and 10943 for ‘test’**. And ‘train-val’ subset is a merge of ‘train’ and ‘val’.
# Folders
## Labels
+ /OPT2017/Annotations: \*.xml
+ /OPT2017/labels: \*.txt *with the form of (class x y width height)*
## Images
+ /OPT2017/JPEGImages: \*.jpg
## Dataset Division
+ /OPT2017/ImageSets/Main: Division of the dataset.
# Statistics
Label|Name|N_Train|N_Val|N_Trainval|N_Test|N_All|Mean Resized Scale /pixel|Resized Scale Std /pixel
:-: |:-: |:-: |:-: |:-: |:-: |:-: |:-: |:-:
1| ship |950|948|18***|1***8|3886|167.44|110.37
2| bridge |1123|1121|2244|2326|4570|246.10|110.53
3| ground track field |859|856|1717|2017|3734|276.50|100.65
4| storage tank |1099|1092|2191|2215|4406|125.60|68.41
5| basketball court |923|920|1843|2033|3876|108.19|57.46
6| tennis court |1043|1040|2083|2212|4295|102.71|38.80
7| airplane |1226|1222|2448|2451|4899|113.21|67.***
8| baseball diamond |1007|1004|2011|2022|4033|231.61|117.85
9| harbor |967|9***|1931|1953|3884|163.96|94.16
10| vehicle |1188|1186|2374|2382|4756|41.96|9.99
11| crossroad |903|901|1804|2219|4023|220.54|59.24
12| T junction |1066|1065|2131|2289|4420|1***.71|54.88
13| parking lot |1241|1237|2478|2480|4958|122.85|54.45
In this table, N_* refers to numbers of objects. 'Train', 'Val', 'Test' are three subsets of the dataset. 'Mean Resized Scale' shows average scale of each category. 'Resized Scale Std' is the standard deviation of category scale.
# FAQ
If any question is met, please contanct me with the e-mail: 1153463027@qq.com.
# Citation
If you find HRRSD dataset useful in your research, please consider citing:
```
@article{zhang2019hierarchical,
title={Hierarchical and Robust Convolutional Neural Network for Very High-Resolution Remote Sensing Object Detection},
author={Zhang, Yuanlin and Yuan, Yuan and Feng, Yachuang and Lu, Xiaoqiang},
journal={IEEE Transactions on Geoscience and Remote Sensing},
volume={57},
number={8},
pages={5535--5548},
year={2019},
publisher={IEEE}
}
```
近期下载者：
相关文件：
收藏者：