Underground mining safety and productivity are increasingly challenged by deeper extraction, complex geological conditions, confined workspaces, and dynamic production activities. Hazards such as rock instability, hazardous gases, water inrush, equipment failures, and limited visibility restrict the effectiveness of conventional inspection and expose personnel to high-risk underground environments. Intelligent inspection technologies, integrating advanced sensing, mobile robotic platforms, communication systems, and artificial intelligence, provide emerging solutions for continuous monitoring, condition assessment, and early risk warning. This review synthesizes recent advances in intelligent inspection technologies for underground mines across six major targets: rock mass conditions, ground support systems, ventilation systems, haulage systems, personnel safety, and underground water hazards. Representative technologies, practical applications, and current limitations are summarized. Future research directions are discussed with an emphasis on multi-source information integration, autonomous inspection, intelligent decision support, and reliable field deployment. This review aims to support the development of safer, more efficient, and more resilient underground mining operations.

Reliable personnel detection in underground mines is essential for safe intelligent mining, but uneven illumination, dust occlusion, thermal interference, and cross-modal parallax can degrade RGB-T perception. We propose an Environment-Driven Fusion and Optimization Network (EnvFONet), which uses environmental degradation as a unified prior for feature representation, multimodal fusion, and training optimization. First, Dynamic Receptive Field Relaxation (DRFR) adaptively interpolates original and locally smoothed thermal features to balance fine-detail preservation and contextual robustness. Second, an Env-AdaIN-guided spatially gated fusion module applies environment-conditioned residual affine modulation and high-frequency pixel-level weighting to reduce cross-modal statistical drift and parallax-induced boundary artifacts. Third, Environment-Adaptive Relaxation Loss (EARL) adjusts regression supervision for difficult matched samples according to environmental degradation and localization quality. On the self-collected Anshan underground iron-ore mine dataset, EnvFONet achieves 96.8% mAP50, 63.9% mAP75, and 62.8% mAP50:95. On the public LLVIP benchmark, it achieves 94.4% mAP50 and 59.6% mAP50:95. These results show that environment-conditioned fusion and optimization improve RGB-T personnel detection robustness under the evaluated degraded conditions.