A New Learning-Based Boosting in Multiple Classifiers for Color Facial Expression Identification
Dhananjoy Bhakta, Goutam Sarker · Advances in intelligent systems and computing · 2016
A new method for automatic color facial expression identification using multiple classifier classification system has been developed. The system is primarily composed of three classifiers. The same color face features are then trained independently using three different classifiers. Now, a super classification technique integrates the decisions coming from each single classifier which outcomes as a final identified expression. To fuse the conclusion drawn by different classifiers, we apply a new technique of learning-based boosting which improves the complete system performance meaningfully in terms of recognition accuracy.