Automatic image cropping via the novel saliency detection algorithm

Image cropping refers to removing unwanted areas from a photographic or illustrated image. The choice of the best picture window both at capture time and during subsequent processing is normally subjective and a wholly manual task. Effective cropping of images should not only use geometric constraints, but consider the image content as well. It is a challenging problem to cropping an image while taking its content into consideration to preserve salient regions and minimize distortions. The definition of saliency can depend on the specific application being considered. In this paper, we propose a novel saliency detection algorithm based on principal component analysis. And then cropping image via the corresponding saliency map. Extensive experiments on a large variety of natural scenes confirm that our method can effective crop nature images while preserve salient regions.