Robust object recognition besed on color segmentation in robot soccer domain
Liu Li · Jisuanji gongcheng yu sheji · 2004
A technique for object recognition is proposed based on color image segmentation in robot soccer domain. To adapt to the change of light condition, YUV color space is used . In this application, real-time performance and robustness are primary concerns. These requirements are met in two steps: an off-line color classification based on Maximum Likelihood criterion and real-time color segmentation. Experiment results show that the proposed technique is reliable and robust even in presence of changing illumination environment conditions.