Visual Attention Region Determination Using Low-Level Features

摘要

In this study, a visual attention region determination approach using low-level luminance, color, and region features is proposed. First, the contrast map of an image is obtained by computing the contrast between each pixel and its "thresholding" neighboring pixels in eight directions, based on the just noticeable difference (JND) model. Then, the block-based edge map and standard deviation representation are used to generate the region mask of the image. The saliency map of the image is generated by using the contrast map and the region mask of the image. Finally, based on the saliency map of the image, a visual attention region determination scheme is proposed to determine the visual attention regions in the image. Based on the experimental results obtained in this study, the performance of the proposed approach is better than that of three comparison approaches.