Human perception tends to firstly pick attended regions, which correspond to prominent objects in an image. Visual attention region detection simulates the behavior of the human visual system (HVS) and detects regions of interest (ROIs) in the image. In this study, a visual attention region detection approach using low-level texture and object features is proposed. The new and improved (shifted) functions are proposed and used in both the proposed texture and object features to ensure that all attended pixels will be extracted. The proposed approach can generate high-quality spatial saliency maps in an effective manner. As compared with three existing approaches, including Stentiford’s, Zhai/ Shah’s, and Park/Moon’s approaches, the proposed approach has a better performance of extracting ROIs in images and low computational complexity.