A Statistical Approach of Texton Based Texture Classification Using LPboosting Classifier

C. Vivek, S. Audithan
2014 Research Journal of Applied Sciences Engineering and Technology  
The aim of the study in this research deals with the accurate texture classification and the image texture analysis has a voluminous errand prospective in real world applications. In this study, the texton co-occurrence matrix applied to the Broadatz database images that derive the template texton grid images and it undergoes to the discrete shearlet transform to decompose the image. The entropy lineage parameters of redundant and interpolate at a certain point which congregating adjacent
more » ... ting adjacent regions based on geometric properties then the classification is apprehended by comparing the similarity between the estimated distributions of all detail sub bands through the strong LP boosting classification with various weak classifier configurations. We show that the resulted texture features while incurring the maximum of the discriminative information. Our hybrid classification method significantly outperforms the existing texture descriptors and stipulates classification accuracy in the state-of-the-art real world imaging applications.
doi:10.19026/rjaset.7.771 fatcat:7mcachjurjacfmetvni4nfhyua