Abstract
The present paper extended the LBP transitions derived from second-order neighbourhood on to third order neighbourhood LBP TN-LBP and derived transitions on Trapezoid patterns for facial expression classification The TN-LBP forms four Trapezoid Patterns TP i e top left bottom right and top right bottom left So far no researcher carried out work on classification problem based on transitions on third-order neighborhood LBP The present paper derived transitions on the two reciprocal Trapezoids of TN-LBP T-TN-LBP i e top left vs bottom right Each of these Trapezoids on TN-LBP will have five pixies and each of them will have 25 i e 32 patterns The present paper derived transitions on two symmetric T-TN-LBP Based on this facial expression recognition algorithm is built The proposed approach is compared with the existing methodsThis work is licensed under a Creative Commons Attribution 4.0 International License.
Copyright (c) 2014 Authors and Global Journals Private Limited