A Maximum Entropy Segmentation Method Based on Mean Shift Clustering

廖建勇,郭斯羽,黄梓效
DOI: https://doi.org/10.3969/j.issn.1006-9348.2009.09.051
2009-01-01
Abstract:In order to segment images effectively,based on consider image noise elimination and edge maintaining and so on,a new method combining the Mean Shift image smoothing method with maximum entropy segmentation method is proposed. Noise can be removed by smoothing image with Mean Shift method,simultaneously,edge of the image can be remained,thus overcoming the weakness of smooth method.This new method can effectively improve the segmentation effect.
What problem does this paper attempt to address?