19-09-2013, 03:15 PM
Classification of Natural Scene Images Based on Emotion
Classification of Natural .docx (Size: 389.18 KB / Downloads: 16)
Abstract
Emotion in natural scene images plays an important role in the way humans perceive an image. Based on the emotion (happiness, sadness, fear, anger etc.) of any human being the images that are viewed by that person can have a significant impact in a sense that if the person is for example in happy mood and he/she views an image that is pleasing then he/she would have a better sense of attachment towards that image and would not accept an image that depicts sadness as an emotion. Although different people may interpret the same image in different ways, we still can build a universal classification for different emotions.
Introduction
The problem of emotion detection poses interesting questions from a research point of view; for instance: how to model the text for the detection task, what features offer the best prediction/detection power, and to what extent it is even possible to accurately distinguish subjective labels such as emotions from a given source text. To predict emotion, we carry out a fairly traditional machine learning method with the addition of feature selection techniques. Specifically, the experiments here use a set of six basic emotions: happiness, sadness, anger, surprise, fear and disgust.
.
Classification
We used two methods for the classification of the local semantic concepts, k-Nearest Neighbour and Support Vector Machine classifiers. Same classification methods were used in the initial method.
Support Vector Machine classifier
Support Vector Machines (SVM) are based on the concept of decision hyper plane. The SVM finds a linear separating hyper plane with a maximal margin in the higher dimensional space. For our experiments, the LIBSVM package [1] with the radial basis function (RBF) kernel was employed. LIBSVM implements the “one-against-one” approach for multi-class classification. For n = 8 classes there are n(n 1) 2 =28 single classifier and each one trains data from two classes. Each binary classification is considered to be a voting.
Emotion Detection from Natural Scene Image
Emotion modelling evoked by natural scenes is challenging issue. In this paper, we propose a novel scheme for analysing the emotion reflected by a natural scene, considering the human emotional status. Based on the concept of original GIST, we developed the fuzzy-GIST to build the emotional feature space. According to the relationship between emotional factors and the characters of image, L*C*H* colour and orientation information are chosen to study the relationship between human's low level emotions and image characteristics. And it is realized that we need to analyse the visual features at semantic level, so we incorporate the fuzzy concept to extract features with semantic meanings.
Edge direction features
As the second kind of feature we use edge direction histogram. It is computed by grouping the edge pixels which fall into edge directions and counting the number of pixel sin each direction. We are applying the canny edge operator and consider 4 directional edges (horizontal, vertical and 2 diagonals) and 1 non-directional edge. Since our sub regions are arbitrary shaped we need to apply simple mirror padding to extend region to a rectangular area.
Texture features
Texture is another important property of images that helps in the image retrieval. We combine texture features with other visual attribute, because texture on its own does not have the capability of finding similar images. But it can classify textured images from non-textured ones.
Algorithms
Color Moment
Color moments are measures that can be used differentiate images based on their features of color. Once calculated, these moments provide a measurement for color similarity between images. These values of similarity can then be compared to the values of images indexed in a database for tasks like image retrieval. The basis of color moments lays in the assumption that the distribution of color in an image can be interpreted as a probability distribution. Probability distributions are characterized by a number of unique moments (e.g. Normal distributions are differentiated by their mean and variance). It therefore follows that if the color in an image follows a certain probability distribution,