A SET OF INVARIANTS BASED ON MASS/AREA
Yishu Liu · Journal of South China Normal University · 2007
Recognition of distorted images is very important in pattern recognition and computer vision,and finding efficient features which are invariant under some geometrical transforms is the key to solving this problem.This paper concentrates on deriving a novel kind of features,namely invariants based on mass/area,which are invariant with regard to translation,rotation, scale changes and reflection.A set of circles centering on the object centroid are generated first,then for each circle,the ratio of mass(or area) within the circle to the total mass(or area) is calculated,which are invariants based on mass/area.Numerical experiments are carried out and a very high recognition ratio is obtained.