Emotion recognition using color and pattern in textile images

In this paper, a novel method is proposed using color and pattern information for recognizing some emotions included in a textile. Here we use 10 Kobayashi emotion keywords. Our method is composed of feature extraction and classification. For accurate emotion recognition, both color and pattern are extracted from a textile. Then the representative color prototypes of a textile are extracted using color quantization method and pattern is described by wavelet transform followed by some statistical terms. The extracted features are given to the neural network (NN)-based classifiers. The advantages of our method include the following: 1) it is a generalized method to accurately recognize emotions in textile images which used in various application domains; 2) it is a fully automatic method with no manual interaction. To prove these advantages, the experiments are performed on 389 textiles obtained from various application domains such as interior, fashion, and artificial ones. Our method shows the precision of 100% and the recall of 99%, regardless of the specific domain.