In recent years, in-seam seismic exploration was developed by leaps and bounds, which played an important role in the identification of geological abnormal bodies such as small faults and collapse columns in coal seams, and improved the efficiency of mechanized coal mining. In order to improve the detection ability of transmission in-seam wave exploration for small-scale structures, the frequency-space (F-X) domain spatial predictive filtering technology combined with the band-pass filtering method was used to extract the in-seam wave Airy phase and improve the accuracy of transmission in-seam wave energy imaging. Based on the simulation data, the wave field and F-X spectrum characteristics of transmission in-seam wave, direct body wave and random noise were analyzed respectively. It was found that both transmission in-seam wave and direct body wave were significantly different from random noise in F-X domain, and direct body wave was mainly concentrated in low frequency band, while in-seam wave Airy phase was mainly concentrated in high frequency band. Based on the analysis results, a joint filtering method using band-pass filtering and F-X spatial prediction filtering was proposed to extract the transmission in-seam wave Airy phase. Through the simulation data, the in-seam Airy phase extraction test was carried out. The test results showed that the method successfully extracted the high-quality in-seam Airy phase. Using this method, the transmission in-seam wave processing and in-seam wave energy attenuation CT imaging were carried out in Jingfang Coal Industry. The results showed that the signal-to-noise ratio of the in-seam wave record was obviously improved by processing, and the weak transmission in-seam wave Airy phase was extracted with high precision. At the same time, the CT results identified small fractures and improved the detection ability of the transmission in-seam wave.
