Comparative Study of Face Recognition Algorithm on Cohn-Kanade Database

Distinctive machine learning routines are deliberately inspected on a number of databases. After having a lot research in this field, it is examined that LBP features are viable and productive for facial expression recognition. We further developed Boosted-LBP to concentrate the most discriminant LBP feature, and the best recognition performance is achieved by utilizing support Vector Machine classifiers with Boosted-LBP features. Also, we examine LBP features for low-determination facial expression recognition, which is a discriminating issue however occasional tended to in the existing work. We see in our investigations that LBP elements perform steadily and vigorously more than a valuable scope of low resolutions of face pictures, and yield promising outcome in compacted low-resolution video sequence caught in real world environment.