TY - GEN
T1 - Edge detection using inertia-based ant colony systems
AU - Chen, Liang
AU - Guo, Lei
AU - Yang, Ning
AU - Yu, Yong
PY - 2007
Y1 - 2007
N2 - In this paper a new algorithm for edge detection using ant colony search is proposed. First, an initial ants' site determination is imposed at beginning of ACS working that can reduce the ants' number and make the ant close to edge's neighborhood. In edge detection process, some unapparent edges should be regard as a gap in the whole line, or noise changed the continuity of the line, therefore we add inertia-based information to the ant's memory. The ant takes a directional accumulation of inertia information that driving ants to bridge the gap. And bigger inertia, longer gap span. With inertia information applied in ACS, the edge detected has more integral and the experiment approves it.
AB - In this paper a new algorithm for edge detection using ant colony search is proposed. First, an initial ants' site determination is imposed at beginning of ACS working that can reduce the ants' number and make the ant close to edge's neighborhood. In edge detection process, some unapparent edges should be regard as a gap in the whole line, or noise changed the continuity of the line, therefore we add inertia-based information to the ant's memory. The ant takes a directional accumulation of inertia information that driving ants to bridge the gap. And bigger inertia, longer gap span. With inertia information applied in ACS, the edge detected has more integral and the experiment approves it.
UR - http://www.scopus.com/inward/record.url?scp=84876753717&partnerID=8YFLogxK
M3 - 会议稿件
AN - SCOPUS:84876753717
SN - 9781615677214
T3 - International Conference on Artificial Intelligence and Pattern Recognition 2007, AIPR 2007
SP - 320
EP - 325
BT - International Conference on Artificial Intelligence and Pattern Recognition 2007, AIPR 2007
T2 - 2007 International Conference on Artificial Intelligence and Pattern Recognition, AIPR 2007
Y2 - 9 July 2007 through 12 July 2007
ER -