CSST Slitless Spectra: Target Detection and Classification with YOLO

TL;DR

This paper introduces a deep learning framework using YOLO models for direct detection, classification, and analysis of spectral traces in CSST slitless spectroscopy, improving speed and accuracy over traditional methods.

Contribution

It presents a novel end-to-end deep learning approach that unifies detection, classification, and parameter estimation in slitless spectroscopic data analysis.

Findings

High detection precision with YOLOv5

Successful quasar identification

Effective redshift binning

Abstract

Addressing the spatial uncertainty and spectral blending challenges in CSST slitless spectroscopy, we present a deep learning-driven, end-to-end framework based on the You Only Look Once (YOLO) models. This approach directly detects, classifies, and analyzes spectral traces from raw 2D images, bypassing traditional, error-accumulating pipelines. YOLOv5 effectively detects both compact zero-order and extended first-order traces even in highly crowded fields. Building on this, YOLO11 integrates source classification (star/galaxy) and discrete astrophysical parameter estimation (e.g., redshift bins), showcasing complete spectral trace analysis without other manual preprocessing. Our framework processes large images rapidly, learning spectral-spatial features holistically to minimize errors. We achieve high trace detection precision (YOLOv5) and demonstrate successful quasar identification and binned redshift estimation (YOLO11). This study establishes machine learning as a paradigm shift in slitless spectroscopy, unifying detection, classification, and preliminary parameter estimation in a scalable system. Future research will concentrate on direct, continuous prediction of astrophysical parameters from raw spectral traces.