Facial Expression Recognition Based on Gabor Wavelet Phase Features

Gabor wavelet transform usually extracts features using Gabor amplitude features, because the Gabor amplitude reflects the energy spectrum of the image, but the phase information contains rich texture information. This paper proposes a facial expression recognition algorithm using Gabor wavelet phase features. First, facial positioning on the human face, extracting facial sub-graphs, and normalizing the expression features of the sub-graphs. Then, using a multi-orientation and multi-frequency set of 2D Gabor wavelet to transform expression, and combining of Gabor coefficient after the transformation so that it can extract the phase characteristics. With K - nearest neighbor classifier to classify, the experimental results that conducted on the JAFFE database demonstrates its efficiency and validity.