地震映像对现场施测条件的要求较低,非常适合小煤窑采空区的勘探,应用前景十分广泛.传统的地震映像方法通常通过观察岩溶附近绕射波的方法来识别,但常常受制于噪声的影响,探测效果不佳,制约了该项技术的发展.本文提出在选择最佳偏移距,最佳频段的基础上,将地震映像资料经S变换数据重构至频率域,以地震波频率衰减的强弱作为识别物探异常区域的标志,并将其应用至六威线六盘水车站煤窑采空区铁路工程物探中,经过与同测线超高密度电法资料对比,物探异常分布基本一致,且与钻孔验证吻合.通过对地震映像法资料进行S变换等特殊处理,为小煤窑采空区物探技术发展提供了一种新思路. The seismic images have low requirements for on-site testing conditions and are very suitable for the prospecting of goafs in small coal mines.Its traditional seismic imaging methods are usually identified by observing the diffraction near the karst but are often subject to noise The detection effect is poor, which restricts the development of this technology.This paper proposes to reconstruct the seismic image data to the frequency domain by selecting the best offset and the best frequency band based on the seismic wave frequency The strength of attenuation is used as a marker to identify geophysical anomaly regions and applied to the geophysical exploration of railway engineering in the coal mined-out area of ​​Liupanshui railway station. After the comparison with the ultra-high density electrical data of the same survey line, the geophysical anomalies are basically the same, Which is in good agreement with the drilling verification.It provides a new idea for the development of geophysical technology in goafs of small coal mines through special processing such as S-transform of seismic data.