Enhancing Safety of Construction Workers: A Review of Stereo Vision-Based Hazard Detection

A typical construction site consists of multiple workers, heavy machinery with moving parts, and unstable structures. This creates a hazardous environment for workers, posing significant safety risks. Traditional safety measures often fail to provide real-time, comprehensive hazard detection. However recent advances in computer vision have led to much research in using computer vision to improve worker safety. This literature review examines how stereo vision, a branch of computer vision, enhances worker safety by offering precise, real-time hazard detection in dynamic environments. The paper provides an overview of the current state of stereo vision technology, highlighting its advantages—such as depth perception, automation, and cost-effectiveness—as well as its challenges, including privacy concerns and environmental limitations. The review discusses future research directions to address these challenges and inspire further research in this field.