Weak celestial source fringes detection based on multi-task learning and pseudo soft threshold residual denoising network

Ruiqing Yan, Zongyao Yin, Cong Dai, Wengping Qi, Xiaojin Shi, Dan Hu, Dan Wu, Xianchuan Yu

Abstract: Highlights•We detect weak celestial source signals from massive astronomical radio data.•We propose a weak celestial source detection method based on multi-task learning.•We propose a pseudo-soft threshold residual denoising network.•We propose a multi-layer fusion feature pyramid network.•Experimental results show that weak celestial source can be detected efficiently.