Combining Convolutional Neural Networks and Point Process for object detection

This code was used to produce the results shown in:

Mabon, J.; Ortner, M.; Zerubia, J. Learning Point Processes and Convolutional Neural Networks for Object Detection in Satellite Images. Remote Sens. 2024, 16, 1019. https://doi.org/10.3390/rs16061019

If you use this code, we strongly suggest you cite:

@Article{rs16061019,
AUTHOR = {Mabon, Jules and Ortner, Mathias and Zerubia, Josiane},
TITLE = {Learning Point Processes and Convolutional Neural Networks for Object Detection in Satellite Images},
JOURNAL = {Remote Sensing},
VOLUME = {16},
YEAR = {2024},
NUMBER = {6},
ARTICLE-NUMBER = {1019},
URL = {https://www.mdpi.com/2072-4292/16/6/1019},
ISSN = {2072-4292},
DOI = {10.3390/rs16061019}
}

For further details you can refer to:

Jules Mabon. Learning stochastic geometry models and convolutional neural networks. Application to multiple object detection in aerospatial data sets. Signal and Image Processing. Université Côte d'Azur, 2023. English. https://theses.hal.science/tel-04404849

Installation

Conda env

• setup conda environment with conda env create -f env.yml

Paths setup

• setup models and data paths with paths_config.yaml.

DOTA metrics

• to compute metrics install dota devkit in data/ (see installation for more info)

cd data/
git clone https://github.com/CAPTAIN-WHU/DOTA_devkit
cd DOTA_devkit/
swig -c++ -python polyiou.i
python setup.py build_ext --inplace

• configure paths_configs.json as needed
• conda env is provided env.yml, setup using conda env create -f env.yml

Results

Results with model mppe_dota_LMM_0305d on sample data from ADS.

Results details, with right plot showing to data and prior scores correspondence to color

Repo structure

• saved_models: saved models that can be used for inference (use setup in paths_config.yaml)
• env.yml: file defining the conda environment suitable to run the code
• model_configs: config files for models

Usage example

# training (use -o to overwrite existing model)
python main.py -p train mppe_dota_LMM_305d.yaml
# inferrence and evalution
python main.py -p inferval mppe_dota_LMM_305d
# making some figures
python tools/figures_results.py

# compute subscores
python tools/compute_subscores.py
# show subscores resutls
python tools/figures_subscores.py

External resources

• Minitel font (CC0 license)