Fast method for detecting hierarchical field contours
Yongfeng Cao · Guangdian gongcheng · 2005
For pattern recognition, different segmentations of an image under different constrains are often needed. A fast hierarchical contours detection method is proposed. This method calculates the saliency value of all possible field contours. With different thresholds, one-pixel-width close contours of different saliency level can easily be detected. Three processes, watershed transformation, basin dynamics and arc dynamics are integrated into one process so that the redundant calculation is removed. Experimental results prove that the efficiency of the proposed method is 1.25 times higher than that of existed similar methods.