Recognition of plastic surgery faces and the surgery types: An approach with entropy based scale invariant features

Abstract

In recent years, the automatic recognition of face comprises many challenging problems which have experienced much consideration due to several applications in different fields. To solve all the situations like the pose, appearance, and lighting changes, and/or ageing the face recognition does not have many methods. The additional challenges which arise recently are Facial expression because of the plastic surgery. This paper deals with a new approach called Entropy-based Volume SIFT (EV-SIFT) for face recognition in an accurate manner after the plastic surgery. The analogous feature extracts the key points and volume of the scale-space structure for which the information rate is determined. Since the entropy is the higher order statistical feature this provides the least effect on uncertain variations in the face. For classification, the corresponding EV-SIFT features are applied to the Support vector machine. The normal SIFT feature extracts the key points on the basis of contrast of the image whereas the V- SIFT feature extracts the key points on the basis of the volume of the structure. Nevertheless, the EV-SIFT technique provides both the volume and contrast information. Finally, the experimental results demonstrate that the EV-SIFT are found to be better on recognizing the plastic surgery faces. Moreover, the methods are experimentally proven for recognizing the type of plastic surgeries such as Blepharoplasty achieves 98%, Brow lift achieves 97%, Liposhaving achieves 96%, Malar augmentation achieves 85%, Mentoplasty achieves 94%, Otoplasty achieves 99%, Rhinoplasty achieves 99% and Skin peeling achieves 91%.