A generalisable framework for saliency-based line segment detection. Issue 12 (December 2015)

• Record Type: Journal Article
• Authors: Brown, Mark; Windridge, David; Guillemaut, Jean-Yves
• Is Part Of: Pattern recognition. Volume 48:Issue 12(2015:Dec.)
• Page Start: 3993
• Page End: 4011
• Publication Date: 2015-12
• Subjects: Line detection -- Feature detection -- Feature matching -- Registration -- Saliency; Pattern perception -- Periodicals; Perception des structures -- Périodiques; Patroonherkenning; 006.4
• DOI: 10.1016/j.patcog.2015.06.015
• Languages: English
• ISSNs: 0031-3203

Abstract

Abstract: Here we present a novel, information-theoretic salient line segment detector. Existing line detectors typically only use the image gradient to search for potential lines. Consequently, many lines are found, particularly in repetitive scenes. In contrast, our approach detects lines that define regions of significant divergence between pixel intensity or colour statistics. This results in a novel detector that naturally avoids the repetitive parts of a scene while detecting the strong, discriminative lines present. We furthermore use our approach as a saliency filter on existing line detectors to more efficiently detect salient line segments. The approach is highly generalisable, depending only on image statistics rather than image gradient; and this is demonstrated by an extension to depth imagery. Our work is evaluated against a number of other line detectors and a quantitative evaluation demonstrates a significant improvement over existing line detectors for a range of image transformations. Abstract : Highlights: A novel information-theoretic salient line segment detection framework is proposed. A saliency-based filtering formulation enables efficient salient line detection. The approach detects the most discriminative lines avoiding repetitive scene parts. The framework is generalisable as demonstrated by an extension to depth imagery. Superior repeatability is demonstrated compared to existing line detectors.