A contiguity-enhanced k-means clustering algorithm for unsupervised multispectral image segmentation

Proceedings of SPIE - The International Society for Optical Engineering

Volumn 3159, Issue , 1997, Pages 108-118

  Theiler, James ^a,b,c   Gisler, Galen ^a,b,c  

^a NIS 2

^b Nonproliferation and International Security Division   (United States)

^c LOS ALAMOS NATIONAL LABORATORY   (United States)

Author keywords

Algorithm; Clustering; Image segmentation; K means

Indexed keywords

ALGORITHMS; IMAGE ENHANCEMENT; K-MEANS CLUSTERING; NATIONAL SECURITY; OPTICAL DATA PROCESSING; PIXELS; REMOTE SENSING; SPECTROSCOPY; STRAIN MEASUREMENT;

CLUSTERING; IMAGE SEGMENTATION ALGORITHM; K-MEANS; MULTI-SPECTRAL DATA; MULTISPECTRAL IMAGES; REMOTE SENSING APPLICATIONS; SPATIAL INFORMATIONS; SPECTRAL INFORMATION;

IMAGE SEGMENTATION;

EID: 0000981892     PISSN: 0277786X     EISSN: 1996756X     Source Type: Conference Proceeding    
DOI: 10.1117/12.279444     Document Type: Conference Paper

References (31)

1. D. M. Titterington, "Updating a diagnostic system using unconfirmed cases, " Appi. Statist. 25, pp. 238-247, 1976.
2. T. J. O'Neil, "Normal discrimination with unclassified observations, " J. Amer. Statist. Assoc. 73, pp. 821-826, 1978.
3. G. J. McLachlan and S. Ganesalingam, "Updating the discriminant function on the basis of unclassified data, " Communications in Statistics-Simulation and Computation 11, pp. 753-767, 1982.
4. C. Ji and G. Boverman. personal communication.
5. V. Castelli and T. M. Cover, "On the exponential value of labeled samples, " Pattern Recognition Lett, 16, pp. 105-111, 1995.
6. R. A. Schowengerdt, Techniques for Image Processing and Classification in Remote Sensing, Academic Press, Orlando, 1983.
7. A. D. Stocker and A. P. Schaum, "Application of stochastic mixing models to hyperspectral detection problems, " in Algorithms for Multispectral and Hyperspectral Imagery III, A. Iverson and S. S. Shen, eds. , Proc SPIE 3071, 1997.
8. J. M. Coggins and A. K. Jam, "A spatial filtering approach to texture analysis, " Pattern Recognition Lett. 3, pp. 195-203, 1985.
9. A. K. Jam and R. C. Dubes, Algorithms for Clustering Data, Prentice Hall, Englewood Cliffs, NJ, 1988.
10. W. M. Rand, "Objective criteria for the evaluation of clustering methods, " J. Amer. Statist. Assoc 66, p. 846, 1971.
11. The results we show here are based on an initial segmentation in which the i'th vertical stripe of the image is assigned to class i mod K. This provides a modicum of initial contiguity (D = 0. 5), without too strongly biasing the cluster formation.
12. K. C. Gowda, "A feature reduction and unsupervised classification algorithm for multispectral data, " Pattern Recognition 17, pp. 667-676, 1984.
13. J. MacQueen, "On convergence of k-means and partitions with minimum average variance, " Ann. Math. Statist. 36, p. 1084, 1965. (Abstract).
14. J. MacQueen, "Some methods of classification and analysis of multivariate observations, " in Proceedings of Fifth Berkeley Symposium on Mathematical Statistics and Probability, L. M. L. Cam and J. Neyman, eds. , pp. 281-297, University of California Press, (Berkeley) , 1967.
15. E. Forgy, "Cluster analysis of multivariate data: efficiency versus interpretability of classifications, " Biometrics 21, p. 768, 1965. (Abstract).
16. P. Montolio, A. Gasull, E. Monte, L. Torres, and F. Marques, "Analysis and optimization of the k-means algorithm for remote sensing applications, " in Pattern Recognition and Image Analysis, N. P. de la Blanca, A. Sanfeliu, and E. Vidal, eds. , vol. 1 of Series in Machine Perception and Artificial Intelligence, pp. 155-170, World Scientific, (Singapore), 1992.
17. H. Späth, Cluster Dissection and Analysis: Theory, FORTRAN programs, Examples, Ellis Horwood Limited, Chichester, UK, 1985.
18. Actually, the computation requires 0(d) time where d is the number of dimensions, but we are treating d as if it were of order one-Treating, but not assuming d = 0(1), since d multiplies all the other times as well, so the comparisons between them are all still valid.
19. If one wants to move points according to the 0 criterion, instead of the nearest center criterion, then the safe radius is slightly less than half the distance to the next cluster. If nj is the number of samples in the current cluster, and k the the number in the nearby cluster, then the safe radius is given by (/T-1)/ c 1/2-e/8 where (n + flk)/[ flk(Thj-1)]
20. G. H. Bull and D. J. Hall, "ISODATA-A novel method of data analysis and pattern classification, " tech. rep. , Stanford Research Institute, Menlo Park, CA, 1965.
21. F. R. Fromm and R. A. Northhouse, "Class: A nonparametric clustering algorithm, " Pattern Recognition Lett. 8, pp. 107-114, 1976.
22. R. Dubes and A. K. Jam, "Clustering techniques: The user's dilemma, " Pattern Recognition Lett. 8, pp. 247-260, 1976.
23. G. A. Carpenter and S. Grossberg, "The ART of adaptive pattern recognition by a self-organinzing neural network, " Computer 21, pp. 77-88, 1988.
24. S. Roberts, G. Gisler, and J. Theiler, "Spatio-spectral image analysis using classical and neural algorithms, " in Smart Engineering Systems: Neural Networks, Fuzzy Logic, and Evolutionary Programming, C. H. Dagli, M. Akay, C. L. P. Chen, B. R. Fernández, and J. Ghosh, eds. , vol. 6 of Intelligent Engineering Systems Through Artificial Neural Networks, pp. 425-430, ASME Press, (New York), 1996.
25. M. Goldberg, D. Goodenough, and S. Shlien, "Classification methods and error estimation for multispectral scanner data, " in Proc. Third Canadian Symposium on Remote Sensing, pp. 125-143, 1975.
26. R. M. Gray, "Vector quantization, " IEEE ASSP Mag. 1, pp. 4-29, 1984.
27. W. H. Equitz, "A new vector quantization clustering algorithm, " IEEE Trans. Acoust. Speech, Signal Processing 37, pp. 1568-1575, 1989.
28. L. M. Ni and A. K. Jam, "A VLSI systolic architecture for pattern clustering, " IEEE Trans. on Pattern Anal. and Mach. Intel. PAMI 7, pp. 80-89, 1985.
29. E. L. Zapata, R. Doallo, F. F. Rivera, and M. A. Ismail, "A VLSI systolic arthitecture for fuzzy clustering, " Microprocessing and Microprogramming 24, pp. 647-654, 1988.
30. J. C. Tilton and J. P. Strong, "Analyzing remotely sensed data on the massively parallel processor, " in Proceedings Seventh International Conference on Pattern Recognition, (Montreal), 1984.
31. F. F. Rivera, M. A. Ismail, and E. L. Zapata, "Parallel squared error clustering on hypercube arrays, " J. Parallel and Distributed Computing 8, pp. 292-299, 1990.