Abstract: Human can see emotion as associated with mood, temperament, personality and disposition. To detect emotions is easy for humans but it’s quite difficult for computer as world is three dimensional but Computer has only two dimensions. Computer seeks to emulate the human emotions by digital image analysis. Humans can use vision to identify objects quickly and accurately. Human can detect emotion using voice based on different parameters like tone, pitch, pace and volume; but in case of digital images detecting emotion just by analysing images is a novel way.
This algorithm has two major parts. First, Template database generation and another is emotion detection. In this, we first extracts face from an image using some basic image processing operations and color models in it. Here we define thresholds to separate our region of interest. Then we perform lip detection on cropped face and this extracted lips are stored in our database with emotion name in database generation phase while in emotion detection phase this extracted lips are compared with series of stored template in database and on the basis of best correlated template emotion is recognized. This method of detecting emotions is simple and fast as compared to previous methods i.e. using brain activity or by speech. Size of database will affect the effectiveness of the proposed algorithm.
Keywords: Emotion Detection, Face Detection, Lip Detection, Skin Segmentation.
Title: Emotion Detection Algorithm
Author: Namrata Mahajan, Harshad Mahajan
International Journal of Electrical and Electronics Research
ISSN 2348-6988 (online)
Research Publish Journals
