International Journal of Scientific & Engineering Research, Volume 7, Issue 2, February-2016 130 ISSN 2229-5518 Emotion Recognition using discrete cosine transform and discrimination power analysis Shikha Surendran Dept. of Electronics and Communication MIT, Anjarakandy, Kannur Kerala
[email protected] ABSTRACT 1. INTRODUCTION The paper states the technique of recognizing A persons face changes according to emotion using facial expressions is a central emotions or internal states of the person. Face is a element in human interactions. Discrete cosine natural and powerful communication tool. transform (DCT) is a powerful transform to Analysis of facial expressions through moves of extract proper features for face recognition. The face muscles leads to various applications. Facial proposed technique uses video input as input expression recognition plays a significant role in image taken through the webcam in real time on Human Computer Interaction systems. Humans which discrete cosine transform is performed. can understand and interpret each other’s facial After applying DCT to the entire face images, changes and use this understanding to response some of the coefficients are selected to construct and communicate. A machine capable of feature vectors. Most of the conventional meaningful and responsive communication is one approaches select coefficients in a zigzag manner of the main focuses in robotics. There are many or by zonal masking. In some cases, the low- other areas which benefit from the advances in frequency coefficients are discarded in order to facial expression analysis such as psychiatry, compensate illumination variations. Since the psychology, educational software, video games, discrimination power of all the coefficients is not animations, lie detection or other practical real- the same and some of them are discriminant than time applications.