Real-time N-finder processing algorithms for hyperspectral imagery
详细信息 查看全文
- 作者：Chao-Cheng Wu (1) chaowu1@umbc.edu
Hsian-Min Chen (23) hsmin6511@gmail.com
Chein-I Chang (14) cchang@umbc.edu - 关键词：N ; FINDR – ; Real ; time circular N ; FINDR (RT Circular N ; FINDR) – ; RT iterative N ; FINDR (RT IN ; FINDR) – ; Real ; time SeQuential N ; FINDR (RT SQ N ; FINDR) – ; Real ; time SuCcessive N ; FINDR (RT SC N ; FINDR) – ; Virtual dimensionality (VD)
- 刊名：Journal of Real-Time Image Processing
- 出版年：2012
- 出版时间：June 2012
- 年：2012
- 卷：7
- 期：2
- 页码：105-129
- 全文大小：2.2 MB
- 参考文献：1. Winter, M. E.: N-finder: an algorithm for fast autonomous spectral endmember determination in hyperspectral data. In: Image Spectrometry V, Proceedings of SPIE, vol. 3753, pp. 266–277 (1999)
2. Boardman, J.W.: Geometric mixture analysis of imaging spectrometery data. Proc. Int. Geosci. Remote Sens. Symp. 4, 2369–2371 (1994)
3. Chaudhry, F., Wu, C., Liu, W., Chang, C.-I., Plaza, A.: Pixel purity index-based algorithms for endmember extraction from hyperspectral imagery, chap. 3. In: Chang, C.-I. (ed.) Recent Advances in Hyperspectral Signal and Image Processing. Research Signpost, Trivandrum (2006)
4. Chang, C.-I., Plaza, A.: Fast iterative algorithm for implementation of pixel purity index. IEEE Geosci. Remote Sens. Lett. 3(1), 63–67 (2006)
5. Plaza, A., Chang, C.-I.: Impact of initialization on design of endmember extraction algorithms. IEEE Trans. Geosci. Remote Sens. 44(11), 3397–3407 (2006)
6. Chang, C.-I., Wu, C.-C.: Random pixel purity index algorithm. IEEE Trans. Geosci. Remote Sens. Lett. (to appear)
7. Chang, C.-I.: Hyperspectral Imaging: Techniques for Spectral Detection and Classification. Dordrecht: Kluwer Academic/Plenum Publishers (2003)
8. Chang, C.-I., Du, Q.: Estimation of number of spectrally distinct signal sources in hyperspectral imagery. IEEE Trans. Geosci. Remote Sens. 42(3), 608–619 (2004)
9. Chang, C.-I., Wu, C., Liu, W., Ouyang, Y.C.: A growing method for simplex-based endmember extraction algorithms. IEEE Trans. Geosci. Remote Sens. 44(10), 2804–2819 (2006)
10. Reed, M., Simons, B. Functional Analysis, Academic Press, New York (1972)
11. http://speclab.cr.usgs.gov/cuprite.html
12. Harsanyi, J.C., Chang, C.-I.: Hyperspectral image classification and dimensionality reduction: an orthogonal subspace projection approach. IEEE Trans. Geosci. Remote Sens. 32(4), 779–785 (1994)
13. Harsanyi, J.C., Farrand, W., Chang, C.-I.: Detection of subpixel spectral signatures in hyperspectral image sequences. In: Annual Meeting, Proceedings of American Society of Photogrammetry and Remote Sensing, Reno, pp. 236–247 (1994)
14. Nascimento, J.M.P., Dias, J.M.: Vertex component analysis: a fast algorithm to unmix hyperspectral data. IEEE Trans. Geosci. Remote Sens. 43(4), 898–910 (2005)
15. Wang, J., Chang, C.-I.: Independent component analysis-based dimensionality reduction with applications in hyperspectral image analysis. IEEE Trans. Geosci. Remote Sens. 44(6), 1586–1600 (2006)
16. Wu, C.C., Lo, C.S., Chang, C.-I.: Improved process for use of a simplex growing algorithm for endmember extraction. IEEE Trans. Geosci. Remote Sens. Lett. 6(3), 523–527 (2009)
17. Wang, J., Chang, C.-I.: Applications of independent component analysis in endmember extraction and abundance quantification for hyperspectral imagery. IEEE Trans. Geosci Remote Sens. 44(9), 2601–2616
18. Winter, E.M., Schlangen, M.J., Hill, A.B., Simi, C.G., Winter, Winter, M.E.: Tradeoffs for real-time hyperspectral analysis. In: Proceedings of SPIE, vol. 4725, Algorithms and Technologies for Multispectral, Hyperspectral, and Ultraspectral Imagery VIII, pp. 366–371 (2002)
19. Wang, J., Chang, C.-I.: FPGA design for real-time implementation of hyperspectral target detection and classification algorithms. In: Plaza, A., Chang, C.-I., (eds.) High-Performance Computing in Remote Sensing, chap. 16, pp. 379–395, Boca Raton: CRC Press (2007) - 作者单位：1. Remote Sensing Signal and Image Processing Laboratory, Department of Computer Science and Electrical Engineering, University of Maryland, Baltimore County, Baltimore, MD 21250, USA2. Department of Radiology, China Medical University Hospital, Taichung, Taiwan, ROC3. Department of Biomedical Engineering, HungKuang University, Taichung, Taiwan, ROC4. Department of Electrical Engineering, National Chung Hsing University, Taichung, Taiwan, ROC
- 刊物类别：Computer Science
- 刊物主题：Image Processing and Computer Vision
Multimedia Information Systems
Computer Graphics
Pattern Recognition
Signal,Image and Speech Processing - 出版者：Springer Berlin Heidelberg
- ISSN：1861-8219
文摘
N-finder algorithm (N-FINDR) is probably one of most popular and widely used algorithms for endmember extraction in hyperspectral imagery. When it comes to practical implementation, four major obstacles need to be overcome. One is the number of endmembers which must be known a priori. A second one is the use of random initial endmembers to initialize N-FINDR, which generally results in different sets of final extracted endmembers. Consequently, the results are inconsistent and not reproducible. A third one is requirement of dimensionality reduction (DR) where different used DR techniques produce different results. Finally yet importantly, it is the very expensive computational cost caused by an exhaustive search for endmembers all together simultaneously. This paper re-designs N-FINDR in a real time processing fashion to cope with these issues. Four versions of Real Time (RT) N-FINDR are developed, RT Iterative N-FINDR (RT IN-FINDR), RT SeQuential N-FINDR (RT SQ N-FINDR), RT Circular N-FINDR, RT SuCcessive N-FINDR (RT SC N-FINDR), each of which has its own merit for implementation. Experimental results demonstrate that real time processing algorithms perform as well as their counterparts with no real-time processing.