22 August 2017 Saliency detection via background invariance in scale space
Author Affiliations +
J. of Electronic Imaging, 26(4), 043021 (2017). doi:10.1117/1.JEI.26.4.043021
Background prior selection is the key step in current ranking-based saliency detection approaches. The existing related methods usually choose boundary regions of an image or the region with low initial saliency value in single image scale as the background. Then, the saliency map is obtained by ranking the inside similarity and correlation. However, these methods cannot handle situations in which the salient object lies on the image boundary (boundary-salient) multiple salient objects exist in a single image (multisalient). To this end, this paper proposes an adaptive background selection method by exploiting the background invariance in different image scales within distinct color spaces. Through embedding the selected background prior into multiple newly proposed ranking-based saliency methods, the superiority of the obtained background prior is strongly verified. Exhaustive experiments on four challenging datasets demonstrate that the proposed method outperforms the state-of-the-art methods in handling the boundary-salient and multisalient situations.
© 2017 SPIE and IS&T
Li Zhou, Yongfeng Ju, Jianwu Fang, Jianru Xue, "Saliency detection via background invariance in scale space," Journal of Electronic Imaging 26(4), 043021 (22 August 2017). https://doi.org/10.1117/1.JEI.26.4.043021 Submission: Received 24 March 2017; Accepted 18 July 2017
Submission: Received 24 March 2017; Accepted 18 July 2017

RGB color model


Magnetic resonance imaging

Image segmentation

Visual process modeling

Detection and tracking algorithms

Information visualization

Back to Top