A Novel CNN-Based Approach for Recognizing Facial Emotion

Emotional recognition based on facial expressions is a very important research topic in the field of Human-Computer-Interaction. This paper proposed a new method to recognize emotions from facial image. First, based on Haar like features to detect left eye, right eye, left eyebrow, right eyebrow, nose and mouth. Then, using convolutional neural network (CNN) to recognize emotions with the above-mentioned detected images as input images. A software system based on the proposed method was designed to experiment. The experimental results with CK+ database and database show that the proposed solution achieved the best recognition rate of 98.1%.