Edge-Based Facial Feature Extraction using Adaptive Canny Operator Edge Detection

Abstract

Facial feature extraction is an essential step in the face detection and facial expression recognition frameworks. To develop a better facial expression recognition system, a good feature extraction method is needed. In this paper, an efficient Facial Feature Extraction method for recognizing four different expressions such as neutral, happy, surprise and sad is presented. In this study, adaptive canny operator edge detection method is used to reduce the computational complexity and improved the accuracy of feature point location. To validate the performance of the proposed feature extraction, the generated features are classified using Maximum Correlation Classifier (MCC). The experimental results demonstrated that the proposed feature extraction method could generate significant facial features and these features are able to be classified into each expression. Our results also showed that the proposed feature extraction method is more efficient than Gabor wavelet edge detection method.