Multi-spectral Texture Characterisation for Remote Sensing Image Segmentation
A multi-spectral texture characterisation model is proposed, the Multi-spectral Local Differences Texem – MLDT, as an affordable approach to be used in multi-spectral images that may contain large number of bands. The MLDT is based on the Texem model. Using an inter-scale post-fusion strategy for image segmentation, framed in a multi-resolution approach, we produce unsupervised multi-spectral image segmentations. Preliminary results on several remote sensing multi-spectral images exhibit a promising performance by the MLDT approach, with further improvements possible to model more complex textures and add some other features, like invariance to spectral intensity.
KeywordsTexture analysis multispectral images Texems
Unable to display preview. Download preview PDF.
- 1.Plaza, A., Martínez, P., Plaza, J., Pérez, R.: Dimensionality reduction and classification of hyperspectral image data using sequences of extended morphological transformations. IEEE Transactions on Geoscience and Remote Sensing 37(6), 1097–1116 (2005)Google Scholar
- 3.Haindl, M., Havlicek, V.: A Simple multispectral multiresolution Markov texture model. In: International Workshop on Texture Analysis and Synthesis, pp. 63–66 (2002)Google Scholar
- 7.Jojic, N., Frey, B., Kannan, A.: Epitomic analysis of appearance and shape. In: IEEE International Conference on Computer Vision, pp. 34–42 (2003)Google Scholar
- 9.Bouman, C.A.: Cluster: An unsupervised algorithm for modelling Gaussian mixtures (April 1997), http://www.ece.purdue.edu/~bouman
- 10.Xie, X., Mirmehdi, M.: Colour image segmentation using texems. Annals of the BMVA 2007(6), 1–10 (2007)Google Scholar