Age group classification based on Bins of Gradients over Gradient Hessianspace facial images

ABSTRACT

Visitors to the museum may have varying levels of interest, language and comprehension depending on their age. Here, a common computer-aided addressing system cannot describe effectively without considering their nature. Hence the efficient assessment of whole facial skin details is an important role in facial age group classification. To address this issue, Bins of Gradients over Gradient Hessianspace (BOGGH) is proposed. Here we first compute gradient space for the detected face and then apply Hessian filter to obtain imprinted edge and curve details. Second, over the gradient Hessianspaced facial images, Bins of Gradients (BOG) technique is applied in multiple directions chosen. Finally, the age features of the facial images are then trained and classified using RBF-based SVM technique. The proposed BOGGH method achieves the classification rate of 99.39% for the FGNET database and 99.72% for the MORPH database which is superior than other existing methods.

KEYWORDS:

• Age group estimation
• gradient features
• gradient space
• Hessianspace
• feature extraction
• SVM classifier
• Bins of Gradients
• radial basis function

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

M. Annalakshmi

M. Annalakshmi received B.E. degree from Thiagarajar College of Engineering in 1998, M.E. degree in Optical Communication from Alagappa Chettiar College of Engineering and Technology in 2005 and Ph.D. in Image Processing from Anna University, Chennai. My research areas are Computer Vision and Pattern Recognition, Object Detection and Classification.

S. Mohamed Mansoor Roomi

S. Mohamed Mansoor Roomi received his B.E. degree from Madurai Kamaraj University, in 1990, his M.E. degree in Power Systems and Communication Systems from Thiagarajar College of Engineering in 1992 and 1997 and his Ph.D. in Image Analysis from Madurai Kamaraj University in 2009. He has authored and co-authored more than 300 papers in various journals and conference proceedings and numerous technical and industrial project reports.