Abstract: The coal mine safety monitoring and control system has many types, different structures, no sharing of information, and no mutual assistance of functions. The monitoring data is difficult to be effectively utilized and coal mine safety production lacks reliable data support. Therefore, a security monitoring system based on micro-service architecture was designed. The information management and automatic monitoring business systems were unified into a flexible, robust and efficient system platform to adapt to big data analysis and mining applications. Through the big data platform of coal mine safety monitoring based on Hadoop, the distributed storage, selective extraction and efficient calculation of massive environmental monitoring data were realized. Through the integration and in-depth mining of production environment monitoring data, the machine learning model was established to ensure the automatic identification of potential safety hazards and recommendation of corresponding treatment measures, thus, providing assistance to the comprehensive research and scientific decision-making of coal mine environmental safety and promoting the intelligent management of coal mine safety.