Abstract: Coal is the “ballast stone” for China’s energy security, but hidden disaster-causing factors and the “three highs and one low” problems in deep underground environments pose threats to the safe, efficient and green mining of coal mines. Currently, the detection of hidden disaster-causing factors in mines mainly relies on geological surveys, drilling and geophysical exploration methods, but these methods have problems such as limited data acquisition, high costs and narrow coverage. Distributed optical fiber sensing technology (DOFS), with its advantages of distributed sensing, long-distance coverage, high-precision measurement and anti-electromagnetic interference, can monitor multiple parameters such as stress, temperature, seepage and vibration in real time, thereby significantly enhancing the detection capability of hidden disaster-causing factors. This paper systematically reviews the core technologies and parameter indicators of DOFS in the monitoring of hidden disaster-causing factors in mines, including fiber Bragg grating (FBG), optical time-domain reflectometry (OTDR), optical frequency-domain reflectometry (OFDR), Brillouin optical time-domain reflectometry (BOTDR), Brillouin optical time-domain analysis (BOTDA), Brillouin optical frequency-domain analysis (BOFDA), Raman optical time-domain reflectometry (ROTDR), Raman optical frequency-domain reflectometry (ROFDR) and distributed acoustic sensing (DAS). A comparative analysis of the above technologies is conducted from multiple perspectives such as monitoring capabilities, applicable scenarios and selection criteria, and typical engineering cases are combined to discuss the application advantages of DOFS in aspects such as the stability analysis of coal mine surrounding rock, health monitoring of support structures, gas disaster early warning, evolution of the temperature field in roadways and state assessment of belt conveyors. The current situation of DOFS technology in precise acquisition of multi-field data, intelligent interpretation of massive data, effective representation of sensing data and joint monitoring with multiple means is analyzed, and optimization schemes are proposed, including clarifying the feedback mechanism between optical fiber sensing data and disaster-causing factors, optimizing the construction of multi-source and multi-field coupling systems, establishing a disaster evolution model based on spatio-temporal big data mining, and promoting the construction of intelligent monitoring and real-time early warning systems. Based on the three-layer architecture of “physical space, sensing space and decision-making space”, this paper proposes a conceptual model of the monitoring and early warning system for hidden disaster-causing factors in mines, and combines DOFS with artificial intelligence technology to achieve intelligent matching and autonomous regulation, thereby providing theoretical and technical support for mine disaster early warning and the construction of smart mines.