Image analysis requires a segmentation step to distinguish the significant components of the image, i.e., the foreground from the background. As a step prior to image classification the quality of the segmentation is of significant importance. High-resolution satellite image classification using standard per-pixel approaches is difficult because of the high volume of data, as well as high spatial variability within the objects. One approach to deal with this problem is to reduce the image complexity by dividing it into homogenous segments prior to classification. This has the added advantage that segments can not only be classified on basis of spectral information but on a host of other features such as neighborhood, size, texture and so forth. Segmentation of the images is carried out using the region based algorithms such as marker-based watershed transform by taking the advantage of multi-resolution and multi-scale gradient algorithms. This paper presents an efficient method for image segmentation based on a multi-resolution application of a wavelet transform and marker-based watershed segmentation algorithm. It also addresses the issue of excessive fragmentation into regions of watershed segmentation, which is avoided by the multi-resolution analysis fact. The most significant components perceived in the highest resolution image will remain identifiable also at lower resolution. Hence the seeds for watershed segmentation on the lower resolution levels are identified and then used them to identify the significant seeds in the highest resolution image. Experimental result of proposed technique gives promising result on QuickBird images. It can be applied to the segmentation of noisy or degraded images as well as reduce over-segmentation.

1. Haralick, R. M.; Shapiro, L. G. (1992): Computer and robot vision. Vol. I, Addison-Weasley, Reading, 672 p.
2. Meinel, G.; Neubert, M. (2004): A Comparision of segmentation programs for high resolution remote sensing data. Int. Arch. of Photogrammetry, Remote Sensing and Spatial Information Sciences XXXV-B4, pp. 1097-1102.
3. Gonzalez, R.C., Woods, R.E., 2002. Digital Image Processing, 2nd ed. Prentice-Hall, Reading, NJ, USA.
4. Pal, N.R., Pal, S.K., 1993. A review on image segmentation techniques. Patt. Recognit. 26 (9), 1277–1294.
5. Bhandarkar, S.M., Hui, Z., 1999. Image segmentation using evolutionary computation. IEEE Trans. Evolut. Comput. 3 (1), 1–21.
6. Kim, H.J., Kim, E.Y., Kim, J.W., Park, S.H., 1998. MRF model based image segmentation using hierarchical distributed genetic algorithm. IEE Electron. Lett. 34 (25), 1394–1395.
7. Kim, J.B., Kim, H.J., 2003. Multi-resolution -based watersheds for efficient image segmentation. Patt. Recogni Lett. 24, 473-488.
8. Beucher, S., Meyer, F., 1993. The morphological approach to segmentation: the watershed transformation. In: Dougherty, E. (Ed.), Mathematical Morphology in Image Processing. Marcel Dekker, New York.
9. Wang, J.Z., Li, J., Gray, R.M., Wiederhold, G., 2001. Unsupervised multi-resolution segmentation for images with low depth of field. IEEE Trans. Patt. Anal. Mach. Intell. 23(1), 85-90.
10. Meyer, F,. and Beucher, S., 1990, “Morphological Segmentation,” Journal of Visual Communication and Image Representation, v. 11, p. 21–46.
11. Liu, J., Yang, Y. H., Multi-resolution color image segmentation, IEEE Trans. Patt. Anal. Intell. 16 (7), 689-700, 1994.
12. Frucci, M., Ramella, G. and Sanniti di Baja, G., 2005, Over segmentation reduction via multiresolution image representation, Pattern Recognition, Image Analysis and Applications, LNCS 3773, pp. 989–996.
13. Lucieer, A., 2004. Uncertainties in Segmentation and Their Visualisation. PhD Thesis, Utrecht University, ITC Dissertation 113, Enschede, 174 p. http://www.itc.nl/library/Papers_2004/phd/lucieer.pdf (Accessed on 18th November 2010)

Multi-resolution Analysis Image Segmentation Watershed Transform High resolution satellite image