Title :
Algorithm of compensating loss of round color mark based on circle fitting
Author :
Zhou, Jun ; Li, Kui ; Wang, Ting-qi
Author_Institution :
Coll. of Mech. & Electr. Eng., Hohai Univ., Jiangsu, China
Abstract :
The color mark recognition is a key step in Mirosot, the recognizing results affect the orientation angle to a great extent. During the capturing process there are some loss of color mark because of the environmental disturbing light, which can lead to a big error by calculation. For the above phenomenon, a new algorithm is designed based on the LSM (least square method), in which a circle is adopted to approximate the edge of the color mark after the pretreatment of the image for extracting the edge of color mark, so as to detect the center and calculate the orientation angle. It has been proved the recognizing precision and the calculating speed can be improved greatly through the experiment.
Keywords :
edge detection; image colour analysis; least squares approximations; mobile robots; multi-robot systems; object recognition; robot vision; circle fitting; color mark recognition; edge extraction; least square method; robot soccer; round color mark; Buffer storage; Color; Educational institutions; Gravity; Image edge detection; Image storage; Least squares approximation; Least squares methods; Robot kinematics; Robot vision systems;
Conference_Titel :
Mechatronics and Automation, 2005 IEEE International Conference
Conference_Location :
Niagara Falls, Ont., Canada
Print_ISBN :
0-7803-9044-X
DOI :
10.1109/ICMA.2005.1626758