Hyperspectral Image Analysis Based on Quad Tree Decomposition
Hyperspectral image analysis is among one of the current trends in computer vision and machine learning. Due to the high dimensionality, large number of classes, presence of noise and complex structure, this is not a trivial task. There exists a need for more precise and computationally efficient algorithms for hyperspectral image segmentation and classification. In this paper, we introduce a novel algorithm for segmentation of hyperspectral images and selecting valuable pixels for classifier training procedure. Our approach is based on Quad Tree Decomposition method, which detects homogeneous region in the input image. This allows to precisely detect groups with similar structure and recognize different possible classes on the image. We discuss the computational complexity of our algorithm and show that it can be easily used in real-life applications. Further, this algorithm is extended by active learning approach, which allows to select a single representative pixel for each detected homogeneous region. With this, the classifier is trained on a significantly reduced dataset without sacrificing its quality. We examine the correlation between the number of folds taken by our segmentation algorithm and used classifiers. We show, that the segmentation procedure can be stopped earlier without drop of the accuracy.
Keywordsmachine learning pattern recognition hyperspectral image image segmentation
Unable to display preview. Download preview PDF.
- 1.Alpaydin, E.: Combined 5 x 2 cv f test for comparing supervised classification learning algorithms. Neural Computation 11(8), 1885–1892 (1999)Google Scholar
- 2.Ceamanos, X., Waske, B., Benediktsson, J.A., Chanussot, J., Fauvel, M., Sveinsson, J.R.: A classifier ensemble based on fusion of support vector machines for classifying hyperspectral data. Int. J. Image Graphics 1(4), 293–307 (2010)Google Scholar
- 3.Chi, M., Kun, Q., Benediktsson, J.A., Feng, R.: Ensemble classification algorithm for hyperspectral remote sensing data. IEEE Geoscience and Remote Sensing Letters 6(4) (2009)Google Scholar
- 4.Chi, M., Qian, Q., Benediktsson, J.A.: Cluster-based ensemble classification for hyperspectral remote sensing images. In: IGARSS (1), pp. 209–212. IEEE (2008)Google Scholar
- 5.Davies, E.R.: Machine Vision: Theory, Algorithms, Practicalities. Elsevier (December 2004)Google Scholar
- 6.Finkel, R., Bentley, J.L.: Quad trees: A data structure for retrieval on composite keys. Acta Informatica (1974)Google Scholar
- 7.Green, R.O., Eastwood, M.L., Sarture, C.M., Chrien, T.G., Aronsson, M., Chippendale, B.J., Faust, J.A., Pavri, B.E., Chovit, C.J., Solis, M., Olah, M.R., Williams, O.: Imaging spectroscopy and the airborne Visible/Infrared imaging spectrometer (AVIRIS). Remote Sensing of Environment 65(3), 227–248 (1998)CrossRefGoogle Scholar
- 8.Samet, H.: Using quadtrees to represent spatial data. NATO ASI Series, vol. F18 (1985)Google Scholar
- 9.Svaetichin, G.: Spectral response curves from single cones. Acta Physiologica Scandinavica. Supplementum 39(134), 17–46 (1956) PMID: 13444020Google Scholar
- 10.Institute of Ophthalmology UCL Colour & Vision Research Laboratories. Colour & vision databaseGoogle Scholar
- 11.Wang, X.-L., Waske, B., Benediktsson, J.A.: Ensemble methods for spectral-spatial classification of urban hyperspectral data. In: IGARSS (4), pp. 944–947. IEEE (2009)Google Scholar