CSST Slitless Spectra: Target Detection and Classification with YOLO

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.

30 pages, 12+6 figures. For high-resolution figures, please see the published version

References in corpus (12)

• Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition
• A Comprehensive Review of YOLO Architectures in Computer Vision: From YOLOv1 to YOLOv8 and YOLO-NAS
• The 2dF Galaxy Redshift Survey: Spectra and redshifts
• The Galactic Center Massive Black Hole and Nuclear Star Cluster
• Galaxia: a code to generate a synthetic survey of the Milky Way
• TPZ : Photometric redshift PDFs and ancillary information by using prediction trees and random forests
• Estimating photometric redshifts with artificial neural networks
• The Slitless Spectroscopy Data Extraction Software aXe
• Deblending and Classifying Astronomical Sources with Mask R-CNN Deep Learning
• Contamination by field late-M, L and T dwarfs in deep surveys
• The many lives of active galactic nuclei-II: The formation and evolution of radio jets and their impact on galaxy evolution
• Galaxy Spectra neural Network (GaSNet). II. Using Deep Learning for Spectral Classification and Redshift Predictions