sabman / CNN-Sentinel

PyCon.DE 2018 talk “Satellite data is for everyone: insights into modern remote sensing research with open data and Python”

Satellite data is for everyone: insights into modern remote sensing research with open data and Python

Video of the talk: https://www.youtube.com/watch?v=tKRoMcBeWjQ

Outline:

• Scripts you will find here
• Requirements (what we used):
• Setup Environment
• How to get Sentinel-2 data
• Datasets

Scripts you will find here

• 01_split_data_to_train_and_validation.py: split complete dataset into train and validation
• 02_train_rgb_finetuning.py: train VGG16 or DenseNet201 using RGB data with pretrained weights on ImageNet
• 03_train_rgb_from_scratch.py: train VGG16 or DenseNet201 from scratch using RGB data
• 04_train_ms_finetuning.py: train VGG16 or DenseNet201 using multisprectral data with pretrained weights on ImageNet
• 04_train_ms_finetuning_alternative.py: an alternative way to train VGG16 or DenseNet201 using multisprectral data with pretrained weights on ImageNet
• 05_train_ms_from_scratch.py: train VGG16 or DenseNet201 from scratch using multisprectral data
• 06_classify_image.py: a simple implementation to classify images with trained models
• image_functions.py: functions for image normalization and a simple generator for training data augmentation
• statistics.py: a simple implementation to calculate mormalization parameters (i.e. mean and std of training data)

Addtionally you will find the following notebooks:

• Image_functions.ipynb: notebook of image_functions.py
• Train_from_Scratch.ipynb: notebook of 05_train_ms_from_scratch.py
• Transfer_learning.ipynb: notebook of 02_train_rgb_finetuning.py

Requirements (what we used):

• python 3.6.6
• tensorflow-gpu (1.11) with Cuda Toolkit 9.0
• keras (2.2.4)
• scikit-image (0.14.1)
• gdal (2.2.4) for 06_classify_image.py

Setup Environment

Append conda-forge to your Anaconda channels:

conda config --append channels conda-forge

Create new environment:

conda create -n pycon scikit-image gdal tqdm
conda activate pycon 
pip install tensorflow-gpu
pip install keras

(or use tensorflow version of keras, i.e. from tensorflow import keras)

See also:

• Keras: https://keras.io/

How to get Sentinel-2 data

1. Register at Copernicus Open Access Hub or EarthExplorer
2. Find your region
3. Choose tile(s) (→ area) and date
   • Less tiles makes things easier
   • Less clouds in the image are better
   • Consider multiple dates for classes like “annual crop”
4. Download L1C data
5. Decide of you want to apply L2A atmospheric corrections
   • Your CNN might be able to do this by itself
   • If you want to correct, use Sen2Cor
6. Have fun with the data

Resources

This talk:

• EuroSAT Data (Sentinel-2, Link)

Platforms for datasets:

• HyperLabelMe: a Web Platform for Benchmarking Remote Sensing Image Classifiers (Link)
• GRSS Data and Algorithm Standard Evaluation (DASE) website (Link)

Datasets:

• ISPRS 2D labeling challenge (Link)
• UC Merced Land Use Dataset (Link)
• AID: A Benchmark Dataset for Performance Evaluation of Aerial Scene Classification (Link)
• NWPU-RESISC45 (RGB, Link)
• Zurich Summer Dataset (RGB, Link)
• Note: Many German state authorities offer free geodata (high resolution images, land use/cover vector data, ...) over their geoportals. You can find an overview of all geoportals here (geoportals)

Image Segmentation Resources:

• Image segmentation with Keras including pretrained weights (Keras-FCN)
• Great link collection of image segmantation networks and datasets (Link)
• Free land use vector data of NRW (BasisDLM or openNRW)

Other:

• DeepHyperX - Deep learning for Hyperspectral imagery: https://gitlab.inria.fr/naudeber/DeepHyperX/