摘要：Computational model of visual attention has got more and more attention in machine vision and image processing. A hierarchical computational model for selective visual attention is proposed in this paper. This model simulates the attention mechanism from far (coarse) to near (fine) of human visual system. Firstly, the input image is analyzed at the coarsest resolution and visual saliency of each part is computed at this level. Regions of attention are selected according to the saliency. Then the sub-regions of the selected region compete for attention at a finer resolution. This process is done iteratively until every salient region and its sub-regions have been processed at different levels respectively. The proposed model has been tested on many natural images. Experiment results show that the proposed model is valid and the attention results are consistent with human visual system.
