Emotional intelligence has important social significance and literature indicates that facial features are an important factor in determining the emotional state. It has been an intense study field to build systems that are capable to recognize emotions automatically based on facial expressions. Various approaches have been proposed but still there is a scope of improvement in detection accuracy because of diverse form of expressions exhibiting the same emotion. A widely used approach in the object detection field is Histogram of Oriented gradients. In this paper extensive experiments are conducted using various subsection sizes of images of histogram of oriented gradients and also along with Local Binary Pattern to extract the features for classification of emotions from facial images. Quantitative analysis of the approach in comparison with others is done to show its applicability and effectiveness.
CITATION STYLE
Kilak, A. S., & Mittal, N. (2017). Classification of emotions from images using localized subsection information. In Communications in Computer and Information Science (Vol. 721, pp. 562–571). Springer Verlag. https://doi.org/10.1007/978-981-10-5427-3_57
Mendeley helps you to discover research relevant for your work.