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

Face Detection and Recognition using Surf and Viola-Jones Algorithm

Author(s):

Jinal R. Thakkar , Kalol Institute & Research Centre, Kalol

Keywords:

Viola-Jones face detector, the cascaded classifier, Face Normalization

Abstract

Face recognition has so many problems like to identify an image robustly of a human face. It also gives databases of some known faces. We use Viola-Jones algorithm for face detection and SURF based approach for face recognition. The principle behind the Viola-Jones algorithm is to scan a sub-window. To rescale the input image to different sizes and then after run the fixed size detector by using these images the standard image processing approach would be. It is time consuming approach because of the calculation of the different size images. On the Contrary by using the standard approach like Viola-Jones detector which is used for rescale instead of the input image and to run the detector many times by using the image – each time with a different size. Both approaches are equally time consuming, but Viola-Jones is a scale invariant detector which requires the same number of calculations of any size. This detector is constructed by using integral image and by using some simple rectangular features of Haar wavelets. This concept represents a scale-invariant and a novel rotation detector and descriptor known as SURF (Speeded-Up Robust Features). SURF is used to perform previously defined schemes with respect to repeatability, robustness and distinctiveness. It is more computing and comparing can be much faster. It is done through on integral images for image convolutions; by making the strengths of the leading existing detectors and descriptors (using a Hessian matrix-based measure for the use of detector, and a descriptor which is distribution-based). It gives result of combination of novel detection and description. This concept shows an overview of the detector and descriptor and then finds out the effects of the most important parameters. This paper uses SURF’s application to two challenging. It achieves such goals like camera calibration which is a special case of image registration and recognition of objects. By this experiment we can say that SURF is very fast and used for image recognition.

Other Details

Paper ID: IJSRDV3I31117
Published in: Volume : 3, Issue : 3
Publication Date: 01/06/2015
Page(s): 1069-1073

Article Preview

Download Article