Change detection in digital imagery using the adaptive learning networks
This paper reports research conducted on the problem of change detection in digital imagery. The detection of changes is very important in any applications which require comparison of many images of the same scene. The problem requires an approach which is flexible and can adapt to varying data trends. The system is based on the adaptive learning networks which are an implementation of the N-tuple method of pattern recognition.
Several experiments were carried out to optimize the net parameters and test the performance of the net for this application. A new mapping structure for the N-tuple was devised to cope with insignificant scattered changes that might occuring a scene. Also the size of the minimum detectable object in a scene was also determined.
KeywordsChange Detection Training Pattern Memory Element Minimum Detectable Change Adaptive Logic
Unable to display preview. Download preview PDF.
- 1-.Shepard, J.R., "A concept of change detection", photogrammetric Eng., vol.30, pp 648–651, July 1964.Google Scholar
- 2-.Lillestrand, R.L., "Techniques for change detection", IEEE Trans. Comput., vol.c-21 pp 654–659, July 1972.Google Scholar
- 3-.Anuta, E. and Bauer, N., "An analysis of temporal data for crop sciences classification and urban change detection", Laboratory for Applications of Remote Sensing, Purdue University, West lafaytte, Indiana, 1973.Google Scholar
- 4-.Eghbal, H.J., "K-S test for detecting changes from Landsat Imagery Data", IEEE Trans. on Systems, Man and Cybernetice, vol.SMC-9, pp 17–23, Jan. 1979.Google Scholar
- 5-.Kawamura, J.G., "Automatic recognition of changes in urban development from aerial photographs", IEEE Trans. Systems, Man, and Cybernetics, vol.SMC-1, pp 230–239, July 1971.Google Scholar
- 6-.Frei, W., Singh, M. and Shibata, T. "Digital Image Change Detection", Optical Engineering vol.19, No.3, pp. 331–337, May–June 1980.Google Scholar
- 7-.Abutaleb, A., Frey, D. and Spicer, K., "Adaptive noise cancelling for change-detection in image sequences", proc. of eleventh Annual Northeast Bioengineering conference, pp 248–251, 1985.Google Scholar
- 8-.Bledsoe, W. and Browning, I. "Pattern recognition and reading by machine", Eastern joint computer conference pp 225–231, 1959.Google Scholar
- 9-.Aleksander, I. and Albrow, R. "Pattern recognition with adaptive logic circuits" IEE/NPL conf. on pattern recognition.Google Scholar
- 10-.Aleksander, I. and Stonham, T.J. "A Guide to Pattern recognition using random access memories" IEEE proceeding on computers and digital techniques, vol.2 no.1, 1979.Google Scholar
- 11-.Aleksander, I. "Microcircuit Learning computers", Mills and Boon Ltd., 1971.Google Scholar
- 12-.Ullmann, J.R. "Experiments with N-tuple method of pattern recognition", IEEE Trans. Computers, vol.c-18, pp.1135–1137, Dec. 1969.Google Scholar
- 13-.Nappey, J.A., "Aspects of N-tuple character recognition for a blind reading aid", Ph.D. Thesis, Brunel University, UK, 1977.Google Scholar
- 14-.Aleksander, I. and Dobree Wilson, M.J., "Adaptive window for image processing" IEE proceedings, vol.132, pt. E. No.5, Sep. 1985.Google Scholar
- 15-.Aleksander, I., "Advanced Digital Information System", Prentice / Hall International, London, 1985.Google Scholar