Vision based construction site surveillance for efficient utilization of resources and safety monitoring

Abstract

My passion lies in computer vision, a field of study which makes computer see and understand what people can. Since humans heavily rely on visual information, building a computer system which has the same visual sense has a significant meaning. With computer vision, computers can smartly interact with the environment and perform various tasks that could only be done by humans before.

The goal of my project is to automatically detect and track workers and equipments from video cameras and localize their position in 3D purely based on computer vision techniques. In this image, 3D position of the workers are displayed on the point cloud model of the construction site. Based on this information, computer can automatically alarm workers walking into a dangerous area. Also, managers can make better decisions about utilization of workers and equipments. I hope small gains in efficiency through my research could lead to an enormous cost savings on the $900 billion construction industry. Awarded first place in the Undergraduate Image of Research Contest 2014. For more information about the Image of Research--Undergraduate Edition go to: http://go.library.illinois.edu/imageofresearch_uredition