The platform mainly aims to improve the safety of mine production, and realizes the safety monitoring and early warning of mine production through the application of deep learning technology. the whole project adopts the combination of service side and edge side. the service side mainly provides mine data analysis and early warning to detect models and data management services. the edge side is based on the jetson platform of vidia to carry out relevant detection tasks in various scenes of the mine, and the detection results are uploaded to the service side. Our Smart Mine program offers several key capabilities to improve mine productivity and safety: Downhole helmet wear detection: Our system can identify whether the miners are wearing the helmet correctly through the camera, so as to ensure the safety when working in a dangerous environment. Monorail Crane Touch Warning: Using advanced image processing technology, our system can alert workers in hazardous areas of the working face of the monorail crane. Data annotation and model training: Users can annotate existing data through our platform to provide valuable information for model training. This allows the model to be continuously optimized and adapted to the specific needs of the mine, which we have now adapted to both scenarios. Instant Alert and Response: Based on cloud-based data analytics and real-time detection at the edge, our system can quickly identify potential problems and issue alerts so that appropriate measures can be taken to reduce the risk of accidents.