BOSC is an innovative tool that leverages AI for advanced agricultural management, promoting sustainability and environmental harmony. Through aerial observation, we are paving the way for a brighter, greener future.
Accurate and efficient label of aerial images is essential for informed decision-making and resource allocation, whether in identifying crop types or delineating land-use patterns. The development of a comprehensive toolbox for manipulating and annotating aerial imagery represents a significant leap forward in remote sensing and spatial analysis.
BOSC functionalities encompass conventional forest management, including tree classification, agricultural monitoring, and applications in semi-urban and urban environment
Ensure you have the following software installed on your machine:
- Python >= 3.7
- (Optional) Miniconda for managing Python environments
- Clone the Repository
First, clone the repository to your local machine using the following command:
git clone https://github.com/RicardDurall/BOSC_toolbox.git
- Create and Activate a Conda Environment (Optional but recommended)
Navigate to the cloned BOSC directory and create a new Conda environment:
conda create -n bosc_env python=3.9
Activate the newly created environment:
conda activate bosc_env
- Install All Dependencies
While inside the BOSC directory and with your Conda environment activated, install the required Python packages using the provided requirements.txt file:
cd BOSC_toolbox
pip install -r requirements.txt
- Download Pretrained Model
Download the pretrained FastSAM model and save the file under the BOSC directory within the cloned repository.
- Launch BOSC
Run the following command to start the toolbox
python app.py
A detailed tutorial can be found in the following video:
Segment Anything and FastSAM provide codes and pre-trained models.
For further questions or collaboration, please do not hesitate to contact us at: ricard.durall.lopez@gmail.com
If you find this project useful for your research, please consider citing us.
@misc{durall2024bosc,
title={BOSC: A toolbox for aerial imagery mapping},
author={Ricard Durall and Laura Montilla and Esteban Durall},
year={2024},
eprint={2406.05833},
archivePrefix={arXiv},
primaryClass={cs.CV}
}