Faster R-CNN object detection model for detecting star-forming clumps in galaxy images from CLAUDS and HSC SSP (U-GRIZY filterbands) using Zoobot for feature extraction.

This model is trained and used to detect star-forming clumps in nearby galaxies observed by CLAUDS and HSC SSP. The model is based on the Faster R-CNN architecture and uses the Zoobot galaxy morphology classification model (https://doi.org/10.21105/joss.05312) as its feature extraction backbone. The model differs also from terrestrial object detection models as it takes 6-channel imaging data as input, U-band from CLAUDS, GRIZY from HSC SSP. The compressed archive contains example code for training the model and running detections on sample galaxy images. Included is also the data used to train the model and the final model weights.