High Impact Factor : 4.396 icon | Submit Manuscript Online icon |

Recognizing Occluded 3d Face Using Block Matching Algorithm and PCA

Author(s):

Riddhi Patel , Gujarat Technological University, Gujarat, India. ; Shruti B. Yagnik, Gujarat Technological University, Gujarat, India.

Keywords:

3D face recognition, Block matching Algorithm, Principal Component Analysis, occlusion.

Abstract

With improvement in sensor technology 3d face recognition has become an imminent technology due to its non co-operative nature. Many researches in 3d face recognition have been dealing with the problem of pose variation, expression, aging and occlusion. We show that under utmost occlusion due to hair, hands, scarves, hates, glass typical 3d face recognition system exhibit poor performance. This paper presents restoration strategy for effective 3d face recognition. In this paper we propose two algorithms for occlusion detection and face recognition. First the occluded face region is detected by matching blocks with the original face. The non occluded regions are then used to restore the missing information. BMA used to restore occluded face. Then apply PCA for recognition. The experimental results show the higher computational efficiency and robustness to occlusion.

Other Details

Paper ID: IJSRDV2I4350
Published in: Volume : 2, Issue : 4
Publication Date: 01/07/2014
Page(s): 483-486

Article Preview

Download Article