An Emotion Based Mental Deceleration Identification Framework

Abstract

Linking the human emotion with the reduced brain skill of that particular person becomes the
social competent concept. Various research methods has been conducted to predict the mental
retardation of people which concludes that the human emotions can be used to predict them
successfully. It is due to uncontrollable emotional states of mentally retarded people
comparatively than the normal mental age people. The mentally retarded people cannot
control their facial emotions which is more difficult to decode. Finding stable emotions of
people can be used to predict the solutions of requirements. There are no research work has
been available to accurately predict the emotional behaviour of humans. The main goal of
this research work is to introduce the system to predict the varying emotional state of people
accurately. This is attained by introducing the new framework namely Emotion based Mental
Retardation Recognition Framework (EMRRF) which can recognize the different kind of
emotions. In this work, input videos are preprocessed first to differentiate the required object
from the noisy pixels and background portions. After preprocessing, feature extraction is
performed to predict the emotions where the extracted features are color, texture and shape
features. The extracted features are learned by applying the Hybridized Particle Swarm
Optimization and Artificial Neural Network (HPSO-ANN) to ensure the accurate prediction
of required object emotional state present in video. The overall experimentation of the
research work is done in the matlab simulation environment from which it is proved that the
proposed research method leads to better result than the existing research works.