- Notifications
You must be signed in to change notification settings - Fork0
MR-DNN: is based on deep learning used to extract rice field from Landsat 8 satellite imagery. Using publicly available satellite imagery data we train a convolutional neural net to predict rice fields in satellite images. MR-DNN achieved tremendous performance for the prediction of rice from Landsat 8 satellite imagery data.
yasir2afaq/Multi-resolution-deep-neural-network
Folders and files
| Name | Name | Last commit message | Last commit date | |
|---|---|---|---|---|
Repository files navigation
MR-DNN: is based on deep learning used to extract rice field from Landsat 8 satellite imagery. Using publicly available satellite imagery data we train a convolutional neural net to predict rice fields in satellite images. MR-DNN achieved tremendous performance for the prediction of rice from Landsat 8 satellite imagery data.
You can download the satellite imagery from the (https://scihub.copernicus.eu/) or USGS Earth Explorer (https://earthexplorer.usgs.gov/). The following are the entity IDs of the images we used. To find images by their ID first select the right dataset (in our case Landsat 8 30m) and then go to "Additional criteria". Here are some IDs as an example we used:
LC08_L2SP_139044_20181001_20200830_02_T1
LC08_L2SP_148039_20190917_20200826_02_T1
- python==3.7.4
- tensorflow-gpu==1.13.1
- keras==2.2.4
- sklearn==0.21.2
- numpy==1.16.4
- matplotlib==3.1.1
- pandas==0.25.1
About
MR-DNN: is based on deep learning used to extract rice field from Landsat 8 satellite imagery. Using publicly available satellite imagery data we train a convolutional neural net to predict rice fields in satellite images. MR-DNN achieved tremendous performance for the prediction of rice from Landsat 8 satellite imagery data.
Resources
Uh oh!
There was an error while loading.Please reload this page.