Abstract: High in-situ stress areas in the coal seam roofs are prone to induce mine dynamic hazard such as rock bursts. To meet the demand for safe coal mining, this study achieved real-time monitoring and swift early warning of high in situ stress areas. Specifically, based on data transmitted from microseismic stations in real time, this study conducted the inversion of the wave velocities in coal seams and rocks using a fast three-dimensional (3D) tomographic inversion algorithm. The accuracy of tomographic inversion results directly determines the judgment on the locations of high in situ stress areas. Therefore, by building a 3D geological model, this study analyzed the influence of different spatial distributions of microseismic events and the spacing between high in situ stress areas and the coal seam roofs on the tomographic inversion results. Furthermore, the method used in this study was tested in a certain mining area. The test results are as follows: (1) Due to different propagation patterns of seismic waves in coal seams and surrounding rocks, the different spatial distributions of microseismic events in coal seams and rocks can reduce the detection accuracy of high in situ stress areas; (2) In the case of too small spacing between high in situ stress areas and the coal seam roofs, the wave velocity gradients near areas with high wave velocities will affect areas with low wave velocities, leading to unclear coal seam locations in the inversion results. (3) The data integrity and inversion accuracy can be effectively improved using evenly distributed microseismic events and rationally arranged microseismic surface observation system. The results of this study will provide a theoretical basis for detecting high in situ stress areas through microseism-based fast 3D tomographic inversion.