The full research and technical details can be found in the article Dynamic Collision Alert System for Collaboration of Construction Equipment and Workers published in Buildings (MDPI).
Because real-world collision data is dangerous and expensive to collect, researchers used a approach:
The video is part of a study that addresses the high rate of accidents in the construction industry. Unlike traditional sensors that fire an alarm whenever any object is near, DCAS uses a to evaluate risk dynamically based on: 999 Part 1(1).mp4
: Recognizes if a worker is facing away or kneeling, which increases risk.
: By using the known size of objects and camera focal lengths, the system can estimate the distance of a worker or machine within a small margin of error. The full research and technical details can be
: The video frames were used to train YOLOv7 (You Only Look Once) and Mask-RCNN models to detect objects and estimate distances accurately in real-time.
: The study noted that moving machine parts (like an excavator's arm) can sometimes obstruct the view or cause perspective distortion, leading to slight distance errors. : By using the known size of objects
: To save time, researchers used the virtual environment to automatically generate bounding boxes around objects, ensuring high precision for the AI training. Key Findings from the Research