Circular Object Recognition Based on Invariant Features
A circular object recognition method utilizing a group of geometric feature parameters is presented.First,Canny operator is used to extract edges and the edge points are traced to form series of sequential sets.All points in each set are fitted to a polygon.Then,the fitted polygon is normalized to obtain invariant features and their corresponding parameters are calculated.In the end,those targets with the feature parameters satisfying certain conditions are recognized as circular objects.Experiments on synthetic images and real images and comparison with RHT method show that the proposed method has the merits of fast recognition rate and high recognition efficiency.
Object Recognition Circle Detection Invariant Feature Polygon Fitting Edge Detection
Aijun Chen
College of Mechanical and Electrical Engineering,Northeast Forestry University,Harbin,Heilongjiang 150040,China
国际会议
大连
英文
233-238
2008-07-27(万方平台首次上网日期,不代表论文的发表时间)