AGE CLASSIFICATION BASED ON FEATURES EXTRACTED FROM THIRD ORDER NEIGHBORHOOD LOCAL BINARY PATTERN

The present paper extended the work carried out by Kumar et. al. [10] on Third order Neighbourhood LBP (TN-LBP) and derived an approach that estimates pattern trends on the outer cell of TN-LBP. The present paper observed and noted that the TN-LBP forms two types of V-patterns on the outer cell of TN-LBP i.e. Outer Right V Patterns (ORVP) and Outer Left V Patterns (OLVP). The ORLP and OLVP of TN-LBP consist of 5 pixels each. The present paper derived Grey Level Co-occurrence Matrix (GLCM) features based on LBP values of ORVP and OLVP. This GLCM is named as ORLVP-GLCM (Outer cell Right and Left V-Patterns of GLCM) and on this four features are evaluated to classify human into child (0 to 12 years), young (13 to 30 years), middle aged (31 to 50 years) and senior adult (above 60 years). The proposed method is experimented on FGNET, GOOGLE and Scanned facial images and the results are compared with the existing methods. The results demonstrate the efficiency of the proposed method over the existing methods.