The detection of deep targets has always been a challenging topic for controlled source electromagnetic methods; however, the time–frequency electromagnetic method (TFEM) can overcome some of these challenges. In this paper, we introduce the method and showcase the technique for imaging deep targets using optimal offsets and transmission periods. In addition, we implement a simulated annealing constrained inversion, based on well-seismic data, to illustrate improvement in the effective detection depth and the ability to identify deep targets. We apply TFEM for two targets in North Xinjiang to show that internal multiple sets of layered series in a deep Carboniferous system and the volcanic rock mass developed in this area can be imaged. We conclude that the low-resistivity layer inside the Carboniferous system is the response of hydrocarbon source beds of muddy clastic rocks and predict that the high-resistivity layer is a favorable reservoir of volcanic rocks. Subsequent drilling confirmed the structural characteristics of the Carboniferous system revealed by TFEM data.