Regions of Interest Detection Algorithm Based on Improved Visual Attention Model

Abstract

Regions of Interest (ROI) detection algorithm based on Visual Attention Model can rapidly focus the attention in the conspicuous target region, and extract the interested region. As to some complicated scenes, it is very difficult to detect the target accurately by using general target detecting method, but using Regions of Interest detection algorithm based on Itti Visual Attention Model can detect the target position very well. But pay attention to Itti Visual Attention Model, it utilizes the luminance, color and texture character of the target to detect the position of it. As little moving target, these characters of it are not obvious, so the detecting result is not satisfactory that utilizing Itti Visual Attention Model directly. According to the problem, this text proposes one Regions of Interest detection algorithm on the basis of improved Itti visual attention model by introducing movement character. The experiment shows that the improved model puts forward a new thinking of little moving target detection.