Face Expression Recognition using Gabor Features and Probabilistic Neural Network

A. Jeyanthi Suresh


This paper describes an improved face emotion recognition (FER) system using Gabor Filter, Probabilistic Neural network (PNN) and Principal Component Analysis (PCA). For face part segmentation and localization, viola jones algorithm is applied. The facial features are extracted from face image by means of Gabor filters. The Probabilistic Neural network (PNN) is used as a classifier for classifying the expressions of supplied face into seven basic categories such as angry, happy, sad, surprise, disgust, fear and neutral. Experiments are conducted on JAFFE facial expression database and gives better performance in terms of 100% recognition rate for training set and 86.2% accuracy for test set. The experiments have highlighted the efficiency of the proposed method in enhancing the classification rate. At the end we have shown simulation results for the proposed technique and established that proposed technique is performing better than the existing work. The planned system is implemented in MATLAB version 8.1.604 R2013a.

Index Terms: Facial expression recognition, Gabor filters, Face regions, Feature extraction, Probabilistic Neural Network

Full Text:


DOI: https://doi.org/10.26483/ijarcs.v6i5.2492


  • There are currently no refbacks.

Copyright (c) 2016 International Journal of Advanced Research in Computer Science