Face Recognition Based on Gradient Integrated Texton Matrix
B. Vamsee Mohan1, V. Vijaya Kumar2

1B.Vamsee Mohan, Rayalaseema University (Research Scholar, Associate Professer, PBR Visvodaya Technology and Science, Kavali, Andhra Pradesh, India. Associate Professor, PBR Visvodaya Technology and Science, B.E, Electronics Communication Engineering College, (Andhra Pradesh), India.
Kumar, Professor, & Dean, Anurag Group of Institutions, Hyderabad (Telangana), India

Manuscript received on 01 May 2019 | Revised Manuscript received on 15 May 2019 | Manuscript published on 30 May 2019 | PP: 2460-2469| Volume-8 Issue-7, May 2019 | Retrieval Number: E3167038519/19©BEIESP
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: In the literature local based methods are popular in extracting facial features. The local binary pattern (LBP) and its variants are one of the popular approaches to extract local features more significantly and precisely and they have been using in many image processing applications. The texons based methods also extract local structural information on a micro grid of 2 x 2 and they are very popular in CBIR. This paper has overcome the disadvantages of the existing texton methods by proposing gradient integrated texton matrix (GITM).This paper initially derives a gradient on facial image and then derives textons on the gradient image. The proposed gradient integrated texton matrix (GITM) defined textons by combining the textons derived in TCM and MTH and GITM has overcome the ambiguity issues of MTH in identification of textons and representation issues of minute textons and complex fusing operations of TCM. The proposed GITM represents the blob, triangle and line shapes of 2 x 2 grids and is specially intended for facial image analysis and can achieve higher face recognition rate than other local based methods. The GLCM features derived on GITM integrates the structural, edge, texture and statistical features of facial images more accurately and precisely. The proposed GITM is specially intended for facial image analysis and can express the spatial correlation of textons and can be considered as a generalized visual attribute descriptor. The proposed GITM descriptor is experimented on popular databases and the results are compared with state of art local representative based methods, the experimental results demonstrate the efficacy of the proposed method over the existing ones.
Keyword: Ocal Features, GLCM, Blob, Triangle, Lines, Structure.
Scope of the Article: Knowledge-based and Expert Systems