Award Date

5-1-2013

Degree Type

Thesis

Degree Name

Master of Science in Computer Science

Department

Computer Science

First Committee Member

Evangelos Yfantis

Second Committee Member

Laxmi Gewali

Third Committee Member

John T. Minor

Fourth Committee Member

Peter Stubberud

Number of Pages

61

Abstract

In construction sites, various guidelines are provided for the correct use of safety equipment. Many fatalities and injuries occur to people because of the lack of exercise of these guidelines and proper monitoring of the violations. In order to improve these standards and amend the cause, a video based monitoring tool will be created for a construction site.

Based on the real time video obtained from cameras on the site, a classification algorithm will be created which has the intelligence to recognize if any safety rules have been violated. A classification vector will be created based on the different classifiers, depending on the properties of the object and the image, to construct a classifier to classify a construction site for a safe state. If any safety rule is being violated the algorithm issues a real time alarm event making the management aware of the violations. The steadiness of the system is indicated by the probability of being in a safe state.

Keywords

Construction industry – Safety measures; Construction site; Image processing; Pattern recognition; Pattern recognition systems; Safety; Safety appliances; Video; Video recording

Disciplines

Computer Sciences

Language

English


Share

COinS