Feature extraction of Heterogeneous face images using SIFT and MLBP algorithm

Humans often use the faces to recognize and similar recognition can enable automatically now by advancement in computing capabilities. The recognition process has now been matured into a science of sophisticated mathematical representation and matching process than early face recognition have been used simple geometric models. Face recognition has received a great deal of attention over the last few years because of its many applications in various domains. The main objective is to extract the distinctive invariance features from the Heterogeneous images from different scenarios that can be use to perform reliable matching between probe and gallery images. The features are highly distinct, so that single feature can be correctly matched with high probability against a large database of features from many images. Previous approaches extracts features from only one type such as matching a face with a gallery but in this paper we are proposing an approach to match a face in different scenario such as near infrared (NIR), thermal photograph, viewed sketch and forensic sketch. Initially we remove the noise from the image. To remove the noise present in the image we use three filters- Gaussian, Difference of Gaussian, CSDN. Then to extract distinct features by using SIFT and MLBP.