Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
532105 | Pattern Recognition | 2014 | 8 Pages |
•We present a new method to detect salient pieces of boundaries in an image.•After detecting meaningful level lines, we label each point as salient/non- salient.•We propose a method to detect meaningful subsequences within these binary sequences.•The method is general and automatic.•The method shows good performance, when tested with different saliency criteria.
A new method to detect salient pieces of boundaries in an image is presented. After detecting perceptually meaningful level lines, periodic binary sequences are built by labeling each point in close curves as salient or non-salient. We propose a general and automatic method to detect meaningful subsequences within these binary sequences. Experimental results show its good performance, when tested with different saliency criteria, such as contrast, regularity, and the combination of both.