Abstract
Superpixel segmentation usually over-segments an image into fragments to extract regional features, thus linking up advanced computer vision tasks. In this work, a novel coarse-to-fine gradient ascent framework is proposed for superpixel-based color image adaptive segmentation. In the first stage, a speeded-up Simple Linear Iterative Clustering (sSLIC) method is adopted to generate uniform superpixels efficiently, which assumes that homogeneous regions preserve high consistence during clustering, consequently, much redundant computation for updating can be avoided. Then a simple criterion is introduced to evaluate the uniformity in each superpixel region, once a superpixel region is under-segmented, an adaptive marker-controlled watershed algorithm processes a finer subdivision. Experimental results show that the framework achieves better performance on detail-rich regions than previous superpixel approaches with satisfactory efficiency.
Highlights
Image segmentation has been widely employed in a wide range of computer vision applications, which is essentially a process of dividing an image into several fragments without intersecting.A superpixel [1] is a homogeneity description of texture, color and other features in accordance with visual sense
Normalized cuts (Ncut) [7] is a representative algorithm based on contour and texture information resulting in regular and compact superpixels, it is poor in accuracy and computing efficiency, especially in dealing with large scale images
The proposed framework is compared with watershed and SLIC [15] to prove the effectiveness, as well as Linear Spectral Clustering (LSC) [16] to demonstrate the superiority
Summary
Image segmentation has been widely employed in a wide range of computer vision applications, which is essentially a process of dividing an image into several fragments without intersecting. Turbopixels [13] adopts level-set based geometric flow for each seed to generate dense over-segmented and compact superpixels It combines a curve evolution model for dilation with a skeletonization process for spatial constraint, but sometimes it provides unsatisfactory results in practice. Linear Spectral Clustering (LSC) [16] adopts a variant of k-means method to iteratively refine uniformly sampled superpixels similar to SLIC Whereas it applies a weighted k-means method in the transformed 10-dimensional feature space by kernel function, which further makes an improvement of Ncut [7]. Jia et al [19] introduce the non-stationarity measure into distance measure and propose nSLIC, which is variable in accordance with local image feature and eliminate the compactness parameter, and eventually improves the visual performance and computing efficiency.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
