A Real Time Algorithm for Planar Shape Recognition and Position Deviation Analysis based on Statistical and Directional Flow-change Methods

J. Yang and X. Chen (Canada)

Keywords

Shape recognition, DFC, position deviation, Mean

Abstract

In this paper, we present a real-time algorithm that can be used to recognize planar shape, estimate the size of the object and analyze the position deviation. Two methods, statistics and directional flow-change (DFC), are combined to develop the algorithm. The statistical method is adopted to estimate the center position and the size of the object, while the DFC method is employed for adjusting the estimated center position and the shape recognition. The experimental results conducted on a vision tracking prototype system show that the algorithm described in this paper is very fast and yields high accuracy.

Important Links:



Go Back