IAPRS & SIS, Vol.34, Part 7, “Resource and Environmental Monitoring", Hyderabad, India,2002
(c)
Fig 4. CBB results with proposed method. The left and right
sides represent images after the stages of SL-1 and SL-2.
Table 4. Summary of image browsing.
Test No. of % of Perfor
obj. Patches after Object detection | mance
Acc.
SL-1 | SL-2 | Correc | False (a-f)/a
stage | stage ta f (in %)
AC-1 23 4 100 13.04 | 86.96
Stadm 11 1 100 0 100
AC-2 20 2 89 11 87.64
4. CONCLUSIONS
The histogram based search scheme followed by evaluation of
the Gabor feature vectors is proposed for browsing the high-
spatial resolution imagery. The suggested combined algorithm
is quite fast and tested for its tolerance against rotation, noise,
location and scale variations. It is found that the tolerance is
rather limited for the scale difference between the query and
test objects. The proposed algorithm was also applied to the
high resolution satellite imagery for typical objects and found
to yield quite satisfactory results. The application of this
method will be extended to multi-spectral high resolution
imagery and reported elsewhere.
REFERENCES
[1]. Bhanu, B. ^ Image Understanding Research For
Automatic Target Recognition”, IEEE Aerospace & Electronic
Systems Magazine, October 1993.
[2]. Brown, W.M., “A Prospectus for Automatic Target
Recognition”, IEEE Trans. On Aerospace & Electronic System,
vol.25, no 3 May1989.
[3]. Manjunath B.S. & Ma W.Y . “Textural features for
browsing and retrieval of image data.” , IEEE PAMI, Vol.18,
pp. 837-842, Aug. 1996.
[4]. L. Xu. el al, *methods of combining multiple classifier and
their applications to handwriting recognition,” IEEE Trans.
System Man Cybernetics, vol. 22, pp. 418-434, 1992.
[5]. Kumar, A.S., and K.L. Majumder,” Information Fusion in
decision tree classification, Int. JI. Of Remote Sens. 22, 861-
869, 2001.
[6]. Kumar, A. S., Chowdhury, S. , Majumder K.L. “Schemes
for integrating multiple neural networks for multi spectral data
classification”, in: Proceedings of the Indian Conference on
Computer Vision, Graphics and Image Processing (ICVGIP
2000), 134-139, 2000.
[7]. Majumdar Jharna & Dilip Y., “Image registration and
Target Tracking”, Proceedings of Indian Conference on
Computer Vision, Graphics and Image Processing (ICVGIP
2000), 154 — 157, 2000.