TY - JOUR
T1 - A fast laser motion detection and approaching behavior monitoring method for moving object alarm system (MOAS)
AU - Dong, Haiwei
AU - Giakoumidis, Nikolas
AU - Juma, Joseph B.
AU - Tretyakov, Dmitriy A.
AU - Mavridis, Nikolaos
N1 - Copyright:
Copyright 2017 Elsevier B.V., All rights reserved.
PY - 2012
Y1 - 2012
N2 - It is very dangerous for elderly people or workers to be hit or even knocked down from fast moving object moving behind them. To avoid such circumstances, we want to build a Moving Object Alarm System (MOAS). This paper focuses on obtaining a fast solution to monitor the moving behavior of the objects by laser for MOAS. Compared with previous work in motion detection and tracking, we are interested especially in monitoring the approaching behavior of such objects, and quickly providing an alarm in the case the situation is dangerous. A boundary within which the objects are monitored was defined. Under this assumption, fan-shaped grid was chosen to get special partitioning. Based on our algorithm, continuous objects can be detected with very high efficiency. By updating a deviation matrix, the object association solution can also be quickly found. Three categories of outdoors experiments were performed: objects passing, approaching as well as crossing, which empirically verified the effectiveness of our proposed method.
AB - It is very dangerous for elderly people or workers to be hit or even knocked down from fast moving object moving behind them. To avoid such circumstances, we want to build a Moving Object Alarm System (MOAS). This paper focuses on obtaining a fast solution to monitor the moving behavior of the objects by laser for MOAS. Compared with previous work in motion detection and tracking, we are interested especially in monitoring the approaching behavior of such objects, and quickly providing an alarm in the case the situation is dangerous. A boundary within which the objects are monitored was defined. Under this assumption, fan-shaped grid was chosen to get special partitioning. Based on our algorithm, continuous objects can be detected with very high efficiency. By updating a deviation matrix, the object association solution can also be quickly found. Three categories of outdoors experiments were performed: objects passing, approaching as well as crossing, which empirically verified the effectiveness of our proposed method.
KW - Approaching behavior
KW - Moving object detction
KW - Object association
UR - http://www.scopus.com/inward/record.url?scp=84901046921&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84901046921&partnerID=8YFLogxK
U2 - 10.1016/j.proeng.2012.07.239
DO - 10.1016/j.proeng.2012.07.239
M3 - Conference article
AN - SCOPUS:84901046921
SN - 1877-7058
VL - 41
SP - 749
EP - 756
JO - Procedia Engineering
JF - Procedia Engineering
T2 - 2nd International Symposium on Robotics and Intelligent Sensors 2012, IRIS 2012
Y2 - 4 September 2012 through 6 September 2012
ER -