An Application Study on Facial Expression Recognition Based on Gabor Wavelets

Abstract

The expression of Gabor wavelet filter is provided, it is explored in detail. In according to actual demand, a new multichannel filter based Gabor wavelet is designed based on theory and practicality. Its center frequency is the range from low frequency to high frequency, its orientation is 6 and scale is 6. It can extract the feature of low quality facial expression image target, and have good robust for automatic facial expression recognition. Experimental results show that the performance of the proposed method is excellent when it is applied to facial expression recognition system. Nowadays, there has been a growing interest in improving aspects of the interaction between humans and computers. It is argued that the facial expressions play an essential role in social interactions with other human beings. Facial expression is a major way of human emotional communication. It is a visible and mutative manifestation of human cognitive activity and psychopathology. It is reported that facial expression constitutes 55% of the effect of a communicated message while language and voice constitute 7% and 38% respectively. With the rapid development of computer vision and artificial intelligence, facial expression recognition becomes the key technology of advanced human computer interaction. More and more people have been paying attention to expression recognition. The research objective of facial expression recognition is how to automatically, reliably, efficaciously use its conveying information. It is a typical issue in model- identification that the automatic recognition system's property is decided by the represented facial expression feature. Therefore, the feature extraction is very important to the facial expressions recognition process. If inadequate features are provided, even the best classifier could fail to achieve accurate recognition. In most cases of facial expression classification, the process of feature extraction yields a definitively large number of features and subsequently a smaller sub-set of features needs to be selected according to some optimality criteria. Gabor filters have been proved to be effective for expression recognition because of its superior capability of multi-scale representation. Gabor wavelet can use very better description of biological visual neuron about receptive field, .According to the needs of special vision, it can adjust the spatial and frequency properties to face expression characteristic wanted, so Gabor filter wavelet is suitable for people face analysis and treatment of expression. In this paper, we pay attention to extract features useful for classification and recognition. The object is the static image. we can obtain the static image utilizing the video tools. The method is simple. It can reliably extract the typical feature and acquire the higher recognition rate. We utilize the responses of Gabor filters which is six orientations and six scales. Experimental results show that the performance of the proposed method is excellent when it is applied to automatic facial expression recognition system. The remainder of this paper is organized as follows: Section 2 of the paper describes Gabor filter's principle, property and the feature characterization in detail. Then the adaptation scheme for choosing the orientation and frequency of Gabor filter to extract the facial expression feature will be performed. The convolution output of the original image is also presented in Section 2. In Section 3,some experimental results are shown and explained. Finally, conclusions and future work are given in Section 4.