Saliency detection using joint spatial-color constraint and multi-scale segmentation

• Published in: Journal of visual communication and image representation Vol. 24; no. 4; pp. 465 - 476
• Main Authors: Xu, Linfeng, Li, Hongliang, Zeng, Liaoyuan, Ngan, King Ngi
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier Inc 01.05.2013; Elsevier
• Subjects: Applied sciences; Artificial intelligence; Biological and medical sciences; Color; Color double-opponent; Computer science; control theory; systems; Exact sciences and technology; Eye and associated structures. Visual pathways and centers. Vision; Fundamental and applied biological sciences. Psychology; Human; Human fixation prediction; Multi-scale technique; Pattern recognition. Digital image processing. Computational geometry; Perception; Psychology. Psychoanalysis. Psychiatry; Psychology. Psychophysiology; Region detection; Representations; Saliency model; Segmentation; Segmentation-based method; Similarity; Similarity distribution; Spatial constraint; State of the art; Vertebrates: nervous system and sense organs; Vision; Visual; Visual attention; Saliency model; Human fixation prediction; Spatial constraint; Multi-scale technique; Similarity distribution; Region detection; Visual attention; Color double-opponent; Segmentation-based method; Computer vision; Segmentation; Modeling; Image analysis; Experimental result; Color vision; Multiscale method; Stimulus salience
• ISSN: 1047-3203; 1095-9076
• DOI: 10.1016/j.jvcir.2013.02.007

[Display omitted] ► We define the joint spatial-color constraint to measure pixel-level saliency. ► The spatial constraint is designed to distinguish the difference between “center and surround”. ► The similarity distribution constraint is developed to detect the salient object and its background. ► The multi-scale segmentation technique is proposed to obtain a consistent saliency map. ► The proposed method outperforms the state-of-the-art methods on salient region detection. In this paper, a novel method is proposed to detect salient regions in images. To measure pixel-level saliency, joint spatial-color constraint is defined, i.e., spatial constraint (SC), color double-opponent (CD) constraint and similarity distribution (SD) constraint. The SC constraint is designed to produce global contrast with ability to distinguish the difference between “center and surround”. The CD constraint is introduced to extract intensive contrast of red-green and blue-yellow double opponency. The SD constraint is developed to detect the salient object and its background. A two-layer structure is adopted to merge the SC, CD and SD saliency into a saliency map. In order to obtain a consistent saliency map, the region-based saliency detection is performed by incorporating a multi-scale segmentation technique. The proposed method is evaluated on two image datasets. Experimental results show that the proposed method outperforms the state-of-the-art methods on salient region detection as well as human fixation prediction.