Salient Region Detection Based on Automatic Feature Selection

Salient regions detection in the image is very useful for image processing. It can reduce the computational complexity of the process and save computation resource. Most of the methods compute the saliency based on feature integration theory. But different features have different contribution to the saliency in different images. A new feature integration method is proposed in this paper. This method can select features and compute the weights of the features dynamically by analyzing the effect of different features on the saliency. Some experiment results and qualitative and quantitative analysis are presented in this paper. The experiment results show that the proposed feature integration strategy can select and weight the appropriate features correctly and the detection results are consistent with human visual system.