In recent years, in the construction industry, there have been frequent incidents of people mistakenly entering hazardous areas, resulting in loss of life and property. Existing intelligent wearable devices are unable to accurately identify and provide real-time monitoring and alerts for hazardous areas, and they are also unable to monitor hazardous areas using a virtual 3D visualization platform. To address these issues, this paper proposes a novel intelligent helmet system with 3D real-time monitoring and alerting capabilities. The system uses a miniature camera to collect information about the surrounding environment, uses a speaker to sound an alarm, and provides real-time feedback to the user. In order to accurately classify and identify hazardous areas, we have improved the YOLOv8 network by adding a small target detection enhancement module(STDEM), which improves the overall performance of the model and its ability to detect distant hazardous area signage. In order to establish the connection between the real world and the virtual world, we construct a digital twin(DT) platform with the help of DT technology to achieve 3D real-time monitoring and information interaction. Experimental results show that our method achieves a mean average precision(mAP) of 93.1% on the self-constructed hazardous area signage dataset. Based on DT and intelligent wearable technology, the system achieves accurate detection of hazardous area signage and information interaction between the real and virtual worlds, which has a wide range of applications in the safety and industrial fields.