2024 Project Description

Deep-learning source detection model for interactive visualization tool – CARTA

Supervisors

Hou, Kuan-Chou, Hwang, Yu-Hsuan, Wang, Kuo-Song and Lee, Chin-Fei
Find out more about supervisors on ASIAA website

Task Description and Goals

This project aims to build a prototype of a source detection model based on deep learning methods, such as a convolutional neural network, expected to be implemented into the cutting-edge visualization tool for astronomical images, CARTA. The student is expected to write and train a Python-based machine-learning model to automatically identify the sources (stars or galaxies) in images for the software. Through this project, the student will learn the application of machine learning in scientific research and understand the basics of astronomical data.

Required Background

– Basic knowledge of Python or similar programming languages
– Basic understanding of a convolutional neural network (e.g., mnist_convnet )