Loading repository data…
jensleitloff / repository
Analyzing Sentinel-2 satellite data in Python with Keras (repository of our talks at Minds Mastering Machines 2019 and PyCon 2018)
A transparent discovery signal based on current public GitHub metadata.
This score does not audit code, security, maintainers, documentation quality, or suitability. Verify the repository and its current documentation before adoption.
Overview about state-of-the-art land-use classification from satellite data with CNNs based on an open dataset
01_split_data_to_train_and_validation.py: split complete dataset into train
and validation02_train_rgb_finetuning.py: train VGG16 or DenseNet201 using RGB data with
pre-trained weights on ImageNet03_train_rgb_from_scratch.py: train VGG16 or DenseNet201 from scratch using
RGB data04_train_ms_finetuning.py: train VGG16 or DenseNet201 using multispectral
data with pre-trained weights on ImageNet04_train_ms_finetuning_alternative.py: an alternative way to train VGG16 or
DenseNet201 using multispectral data with pre-trained weights on ImageNet05_train_ms_from_scratch.py: train VGG16 or DenseNet201 from scratch using
multispectral data06_classify_image.py: a simple implementation to classify images with
trained modelsimage_functions.py: functions for image normalization and a simple
generator for training data augmentationstatistics.py: a simple implementation to calculate normalization
parameters (i.e. mean and std of training data)Additionally you will find the following notebooks:
Image_functions.ipynb: notebook of image_functions.pyTrain_from_Scratch.ipynb: notebook of 05_train_ms_from_scratch.pyTransfer_learning.ipynb: notebook of 02_train_rgb_finetuning.pyWe have defined the requirements in requirements.txt. We used:
06_classify_image.pyAppend 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:
This talk:
Platforms for datasets:
Datasets:
Image Segmentation Resources:
Other:
Jens Leitloff and Felix M. Riese, "Examples for CNN training and classification on Sentinel-2 data", Zenodo, 10.5281/zenodo.3268451, 2018.
@misc{leitloff2018examples,
author = {Leitloff, Jens and Riese, Felix~M.},
title = {{Examples for CNN training and classification on Sentinel-2 data}},
year = {2018},
DOI = {10.5281/zenodo.3268451},
publisher = {Zenodo},
howpublished = {\href{http://doi.org/10.5281/zenodo.3268451}{http://doi.org/10.5281/zenodo.3268451}}
}
Selected from shared topics, language and repository description—not editorial ratings.
parvvaresh /
The gee library is a Python package designed for processing and analyzing Sentinel-1 and Sentinel-2 satellite imagery using Google Earth Engine (GEE). It allows for the loading, processing, and combining of Sentinel-1 and Sentinel-2 data, as well as the calculation of various vegetation indices.
49/100 healthVedant-Git-dev /
GeoVision is a web application that analyzes satellite imagery to detect land cover changes over time.
64/100 healthsahilmerai /
Analyzing urban green cover changes using satellite imagery and Google Earth Engine (GEE). This project utilizes remote sensing techniques to assess vegetation elasticity, land use changes, and sustainability in urban environments, contributing to SDG 15 (Life on Land) and climate resilience. 🚀🌍
ethylmethylether /
Used Sentinel-2 satellite imagery and Python to analyze vegetation changes in Islamabad Region over a 9-year period. Leveraged NDVI (Normalized Difference Vegetation Index) to quantify vegetation health and track land-cover changes.
42/100 healthSquirrelcoding /
A geospatial analysis project to analyze the use of parking lots using various satellite image datasets (i.e, NAIP, Sentinel-2, etc.) in the US. The goal is to observe how cars are distributed across parking lots throughout the various times of the day and the year.
48/100 healthInfermatt66 /
A Python pipeline for downloading, processing, and analyzing Sentinel-2 satellite imagery to calculate seasonal vegetation indices (NDVI) and statistical metrics (median, standard deviation) for forest classification applications.
52/100 health