HI Source

HI source identification focuses on automatically detecting and characterizing neutral hydrogen (HI) regions in large radio astronomical datasets, crucial for understanding galaxy formation and evolution. Current research emphasizes developing and comparing automated source-finding techniques, employing methods like deep learning architectures (e.g., 3D-Unet, V-Net) and traditional algorithms (e.g., SoFiA, MTObjects), often enhanced by post-processing classifiers such as random forests. These advancements improve the efficiency and accuracy of analyzing vast HI surveys, enabling more comprehensive studies of the HI distribution in the universe. The resulting improved data analysis facilitates a deeper understanding of galactic structure and the interplay between galaxies and their environment.

Papers

March 29, 2024

Automated Identification and Segmentation of Hi Sources in CRAFTS Using Deep Learning Method
Zihao Song, Huaxi Chen, Donghui Quan, Di Li, Yinghui Zheng, Shulei Ni, Yunchuan Chen, Yun Zheng
Deep Learning Segmentation Based Approach Automatic Identification Traditional CrAFT Source Detection HI Source

November 30, 2022

Stochastic Parameterization of Column Physics using Generative Adversarial Networks
B. T. Nadiga, X. Sun, C. Nash
Generative Adversarial Network Climate Model Stochastic Parameterization Atmospheric Science Structural Column HI Source

November 23, 2022

A comparative study of source-finding techniques in HI emission line cubes using SoFiA, MTObjects, and supervised deep learning
J. A. Barkai, M. A. W. Verheijen, E. T. Martínez, M. H. F. Wilkinson
Deep Learning Barzilai Borwein Technique Data Cube HI Source

HI Source

Papers

Automated Identification and Segmentation of Hi Sources in CRAFTS Using Deep Learning Method

Stochastic Parameterization of Column Physics using Generative Adversarial Networks

A comparative study of source-finding techniques in HI emission line cubes using SoFiA, MTObjects, and supervised deep learning