Award Date

December 2015

Degree Type

Thesis

Degree Name

Master of Science in Computer Science

Department

Computer Science

First Committee Member

Evangelos Yfantis

Second Committee Member

Harlod Bergehel

Third Committee Member

Andreas Stefik

Fourth Committee Member

Ying Tian

Number of Pages

47

Abstract

Cameras have become common in our society and as a result there is more video available today than ever before. While the video can be used for entertainment or possibly as storage it can also be used as a sensor capturing crucial information, The information captured can be put to all types of uses, but one particular use is to identify a fall. The importance of identifying a fall can be seen especially in the older population that is affected by falls every year. The falls experienced by the elderly are devastating as they can cause apprehension to normal life activities and in some cases premature death. Another fall related issue is the intentional deception in a business with intent of insurance fraud. Classification algorithms based on video can be constructed to detect falls and separate them as either accidental or intentional. This thesis proposes an algorithm based on frame segmentation, and speed components in the x, y, z directions over time t. The speed components are estimated from the video of orthogonally positioned cameras. The algorithm can discern between fall activities and others like sitting on the floor, lying on the floor, or exercising.

Keywords

Computer Vision; Fall Detection; human activity analysis; Video Surveillance

Disciplines

Computer Sciences

Language

English


Share

COinS