---Automatic Search for Low Surface Brightness Galaxies from SDSS images Using Deep Learning
scipy==1.2.1 numpy==1.17.0 matplotlib==3.1.2 opencv_python==4.1.2.30 torch==1.2.0 torchvision==0.4.0 tqdm==4.60.0 Pillow==8.2.0 h5py==2.10.0
Install all dependencies in bulk by entering pip install -r requirements.txt in the PyCharm terminal.
The LSBG dataset is stored in the Galaxy_data/Galaxy_1 directory, which includes both images and labels.
The default parameters in train.py are used for training the LSBG dataset. Run train.py directly to start the training.
Training weights are saved in the logs directory by default, but this can be modified within train.py.
Prediction of the training results requires two files: centernet.py and predict.py.
First, you need to modify the model_path inside centernet.py.
model_path should point to the trained weights file located in the logs directory.
Then, run predict.py to perform the detection.
By default, it detects images in 13_val.txt. The detection results are saved in the map_out directory. You can modify the predict_path and map_out_path inside predict.py to change the images to be detected and the output directory for the results.
https://github.com/liangzengxu/DACL-LSBG