Enhancing Construction Site Safety Using Computer Vision and Drones

Abstract

Despite established requirements for the use of personal protective equipment (PPE), such as helmets and goggles, traditional methods of control at construction sites often fail to ensure the proper level of compliance with safety standards, resulting in increased injuries. The goal of the project is to develop an integrated system of face control, automatic verification of the presence of PPE, and alcohol level monitoring using computer vision technologies, machine learning, and drones. The system will perform face recognition, control the presence of helmets and glasses, as well as identify unsafe behavior and signs of alcohol intoxication, providing continuous monitoring of workers. It is expected that this will significantly improve the level of safety, reduce the number of accidents, and increase labor productivity. The development will be adaptable for various industries - from construction to mining and oil and gas. The project has high practical and socio-economic significance, contributes to the development of the scientific and technical potential of Kazakhstan, the creation of new jobs in the field of IT and automation, as well as strengthens the interaction between science and production.