People often make decisions based on sensitivity rather than rationality. In the field of biological information processing,\nmethods are available for analyzing biological information directly based on electroencephalogram: EEG to determine the\npleasant/unpleasant reactions of users. In this study, we propose a sensitivity filtering technique for discriminating preferences\n(pleasant/unpleasant) for images using a sensitivity image filtering systembased on EEG.Using a set of images retrieved by similarity\nretrieval,we performthe sensitivity-based pleasant/unpleasant classification of images based on the affective features extracted from\nimages with the maximum entropy method: MEM. In the present study, the affective features comprised cross-correlation features\nobtained from EEGs produced when an individual observed an image. However, it is difficult to measure the EEG when a subject\nvisualizes an unknown image.Thus, we propose a solution where a linear regressionmethod based on canonical correlation is used\nto estimate the cross-correlation features from image features. Experiments were conducted to evaluate the validity of sensitivity\nfiltering compared with image similarity retrieval methods based on image features.We found that sensitivity filtering using color\ncorrelograms was suitable for the classification of preferred images,while sensitivity filtering using local binary patterns was suitable\nfor the classification of unpleasant images. Moreover, sensitivity filtering using local binary patterns for unpleasant images had a\n90% success rate.Thus, we conclude that the proposed method is efficient for filtering unpleasant images.
Loading....