Browse > Article
http://dx.doi.org/10.5391/IJFIS.2016.16.4.299

Adaptive Image Segmentation Based on Histogram Transition Zone Analysis  

Acuna, Rafael Guillermo Gonzalez (Universidad Nacional Abierta y a Distancia de Mexico)
Mery, Domingo (Department of Computer Science, Pontificia Universidad Catolica de Chile)
Klette, Reinhard (Auckland University of Technology)
Publication Information
International Journal of Fuzzy Logic and Intelligent Systems / v.16, no.4, 2016 , pp. 299-307 More about this Journal
Abstract
While segmenting "complex" images (with multiple objects, many details, etc.) we experienced a need to explore new ways for time-efficient and meaningful image segmentation. In this paper we propose a new technique for image segmentation which has only one variable for controlling the expected number of segments. The algorithm focuses on the treatment of pixels in transition zones between various label distributions. Results of the proposed algorithm (e.g. on the Berkeley image segmentation dataset) are comparable to those of GMM or HMM-EM segmentation, but are achieved with significantly reduced computation time.
Keywords
image segmentation; image analysis; comparative evaluation;
Citations & Related Records
연도 인용수 순위
  • Reference
1 P. Krahenbuhl and V. Koltun, Geodesic object proposals, in ECCV, pp. 725-739, 2014.
2 R. Klette, Concise Computer Vision: An Introduction into Theory and Algorithms, Springer London, 2014. [Online]. Available: books.google.com.mx/books?id=ZCu8BAAAQBAJ
3 A. Qin and D. A. Clausi, Multivariate image segmentation using semantic region growing with adaptive edge penalty, IEEE Trans. Image Processing, 19:2157-2170, 2010.   DOI
4 S. Lankton and A. Tannenbaum, Localizing region-based active contours, IEEE Trans. Image Processing, 17:2029-2039, 2008.   DOI
5 P. A. Arbel'aez and L. D. Cohen, A metric approach to vector-valued image segmentation, Int. J. Computer Vision, 69:119-126, 2006.   DOI
6 N. Otsu, A threshold selection method from gray-level histograms, IEEE Trans. Systems Man Cybernetics, 9:62-66, 1979.   DOI
7 W. Feng, J. Jia, and Z.-Q. Liu, Self-validated labeling of Markov random fields for image segmentation, IEEE Trans. Pattern Analysis Machine Intelligence, 32:1871-1887, 2010.   DOI
8 J. Long, E. Shelhamer, and T. Darrell, Fully convolutional networks for semantic segmentation, CVPR, pp. 3431-3440, 2015.
9 J. S. Weszka and A. Rosenfeld, Threshold evaluation techniques, IEEE Trans. System Man Cybernetics, 8:622-629, 1978.   DOI
10 R. Achanta, A. Shaji, K. Smith, A. Lucchi, P. Fua, and S. Susstrunk, SLIC superpixels compared to state-of-theart superpixel methods, IEEE Trans. Pattern Analysis Machine Intelligence, 34:2274-2282, 2012.   DOI
11 W. Rand, Objective criteria for the evaluation of clustering methods, J. American Statistical Association, 66 (336): 846-850, 1971.   DOI
12 6D Vision, www.6d-vision.com/scene-labeling, 2016.
13 S. Valentine The Hidden Power of Blend Modes in Adobe Photoshop Adobe Press, 2012.
14 L. Gupta and T. Sortrakul, A Gaussian-mixture-based image segmentation algorithm, Pattern Recognition, 31:315-325, 1998.   DOI
15 Q. Wang, HMRF-EM-image: implementation of the hidden Markov random field model and its expectationmaximization algorithm, arXiv preprint arXiv:1207.3510, 2012.