Block-Based Stereo Matching Using Image Segmentation 


Vol. 44,  No. 7, pp. 1402-1410, Jul.  2019
10.7840/kics.2019.44.7.1402


PDF
  Abstract

In this paper, we propose the approach to stereo matching based on block-based method and using image segmentation. Block based matching, local method, has been popularly and widely employed for stereo matching. In practice, however, it sometimes does not provide sufficiently accurate matching results if textureless region or occlusion is inherent in a scene of interest. This paper suggests the algorithm that provides higher matching accuracy than the existed ones by segmenting the sub-region of an image to acquire boundary regions followed by searching the matched region in the reference image. Searching process is carried out based on high correlation and the lowest difference between matching areas. Middlebury dataset is used in the experiments and the proposed approach outperforms the existed block-matching algorithm.

  Statistics
Cumulative Counts from November, 2022
Multiple requests among the same browser session are counted as one view. If you mouse over a chart, the values of data points will be shown.


  Cite this article

[IEEE Style]

J. Kim, C. Park, D. Lee, "Block-Based Stereo Matching Using Image Segmentation," The Journal of Korean Institute of Communications and Information Sciences, vol. 44, no. 7, pp. 1402-1410, 2019. DOI: 10.7840/kics.2019.44.7.1402.

[ACM Style]

Ji-su Kim, Cheol-hyeong Park, and Deok-woo Lee. 2019. Block-Based Stereo Matching Using Image Segmentation. The Journal of Korean Institute of Communications and Information Sciences, 44, 7, (2019), 1402-1410. DOI: 10.7840/kics.2019.44.7.1402.

[KICS Style]

Ji-su Kim, Cheol-hyeong Park, Deok-woo Lee, "Block-Based Stereo Matching Using Image Segmentation," The Journal of Korean Institute of Communications and Information Sciences, vol. 44, no. 7, pp. 1402-1410, 7. 2019. (https://doi.org/10.7840/kics.2019.44.7.1402)

Search
(IN TITLE, AUTHOR, ABSTRACT,KEYWORDS)

Advanced Search ▼

POPULAR KEYWORDS
(TOP 10 KEYWORDS)

Recent Publications
(LAST 3 YEARS)






pISSN : 1226 - 4717
eISSN : 2287 - 3880