Superpixel Segmentation Based on Global Similarity and Contour Region Transform

Bing LUO  Junkai XIONG  Li XU  Zheng PEI  

IEICE TRANSACTIONS on Information and Systems   Vol.E103-D   No.3   pp.716-719
Publication Date: 2020/03/01
Publicized: 2019/12/03
Online ISSN: 1745-1361
DOI: 10.1587/transinf.2019EDL8153
Type of Manuscript: LETTER
Category: Image Recognition, Computer Vision
superpixel segmentation,  global similarity,  contour,  SLIC,  

Full Text: PDF>>
Buy this Article

This letter proposes a new superpixel segmentation algorithm based on global similarity and contour region transformation. The basic idea is that pixels surrounded by the same contour are more likely to belong to the same object region, which could be easily clustered into the same superpixel. To this end, we use contour scanning to estimate the global similarity between pixels and corresponded centers. In addition, we introduce pixel's gradient information of contour transform map to enhance the pixel's global similarity to overcome the missing contours in blurred region. Benefited from our global similarity, the proposed method could adherent with blurred and low contrast boundaries. A large number of experiments on BSDS500 and VOC2012 datasets show that the proposed algorithm performs better than traditional SLIC.