Abstract:Detecting the focused region in one image is an importance research direction in image fusion and computer vision areas. Focused region belongs to salient region, but current salient detection algorithms have difficult to perfectly detect focused region. Due to the limitations, in this paper we propose a novel solution based on recently developed RPCA (Robust Principal Component Analysis) technology to efficiently detect the focused region. The main idea is using the sparse matrix decomposed from RPCA result, which contains the characteristic of focused region. The experimental results show the efficiency of detecting the focused region, which point out that the RPCA based method has vast applications.