DEtection and Tracking Algorithm for Coronal Holes (DETACH)

Usage

Prerequisites

Step 0. Initial a conda/virtualenv environment and activate it.

For conda, e.g.

conda create --name detach-env python=3.8 -y
conda activate detach-env

For virtualenv, e.g.

virtualenv detach-env
source detach-env/bin/activate

Step 1. Install PyTorch following official instructions, e.g.

conda install pytorch torchvision -c pytorch

or

pip install torch torchvision

Step 2. Install mmdetection following official instructions, e.g.

pip install openmim
mim install mmengine
mim install "mmcv>=2.0.0"
mim install mmdet
mim install mmpretrain

Step 3. Install other requirement libraries

cd /path/to/detach
pip install -r requirement.txt

Step 4. Move the checkpoint files DETACH_detection.pth to the checkpoints folder and edit the configuration file config.json

Detection & Tracking

Only detect and plot image

python comparing_dataset_prediction.py

Detect, track and build database:

python main.py

Plot result from database:

python plot_result_from_db.py

Training details

For simplicity, not all the details of the training are shown in this repository. For those interested, this repository provides our full training code.