Abstract
In this paper, we intend to implement multi-scale texture segmentation by fractional differential. We propose two fractional differential masks and present the structures and parameters of each mask, respectively, on eight directions. Moreover, by theoretical and experimental analysis, we find the better performance fractional differential mask. Finally, we further discuss the capability of fractional differential for multi-scale texture segmentation. Experiments show that, for rich-grained digital images, the capability for multi-scale texture segmentation by fractional differential-based approach appears efficient.
Acknowledgements
The work is supported by the China Postdoctoral Science Foundation (Grant No. 20060401016), Fondation Franco-Chinoise Pour La Science Et Ses Applications (FFCSA), China National Nature Science Foundation (Grant No. 60972131 and 60572033), and the Doctor Foundation of China National Education Department (Grant No. 20060610021).