This dataset includes the following components: 1. Code of the GalCenterNet: A deep learning-based object detection algorithm designed to identify low surface brightness galaxies from large-scale astronomical image datasets. The GalCenterNet is trained using composite images from the Sloan Digital Sky Survey (SDSS) Data Release 16, these images are retrieved from the SDSS Science Archive Server and have been processed into 3-color images using the i-r-g to R-G-B color mapping method. 2. Catalog of Low Surface Brightness Galaxy Candidates: A catalog containing low surface brightness galaxies identified by the GalCenterNet. 3. Catalog of Low Surface Brightness Galaxies: Catalogs for both the training and validation sets, consist of confirmed low surface brightness galaxies.