Human facial emotion recognition through automatic clustering based morphological segmentation and shape/orientation feature analysis

The objective of this work is to develop a system to recognize the human emotions in terms of his/her facial expressions. The work consists of 3 modules. First to segment the face from the human image. Otsu's thresholding method is used in the third plane of YCbCr image to perform background subtraction. Secondly to segment the parts eye, eye gap, forehead, mouth, and nose max-min mean algorithm is used to select the initial cluster values of k-means algorithm. Then the distance based algorithm is used to accurately segment the left right eye, eye gap, forehead, nose, and mouth. Thirdly different shape based features are extracted for the parts segmented in step 2 for different facial expressions. By applying a two level rule based classifier for the data features extracted, we made the machine to understand the emotion of the human. We are trying to emulate the human visual interpretation into machine through our work.