Taking Fuzzy-Rough Application to Mars
This paper presents a novel application of fuzzy-rough set-based feature selection (FRFS) for Mars terrain image classification. The work allows the induction of low-dimensionality feature sets from sample descriptions of feature patterns of a much higher dimensionality. In particular, FRFS is applied in conjunction with multi-layer perceptron and K-nearest neighbor based classifiers. Supported with comparative studies, the paper demonstrates that FRFS helps to enhance the effectiveness and efficiency of conventional classification systems, by minimizing redundant and noisy features. This is of particular significance for on-board image classification in future Mars rover missions.
KeywordsFeature Selection Hide Node Feature Selection Technique Mars Exploration Rover Panoramic Camera
Unable to display preview. Download preview PDF.
- 1.Castano, R., et al.: Current results from a rover science data analysis system. In: Proc. of IEEE Aerospace Conf. (2006)Google Scholar
- 6.Jensen, R., Shen, Q.: Computational intelligence and feature selection: rough and fuzzy approaches. IEEE Press/Wiley (2008)Google Scholar
- 7.Kachanubal, T., Udomhunsakul, S.: Rock textures classification based on textural and spectral features. Proc. of World Academy of Science, Eng. and Tech. 29, 110–116 (2008)Google Scholar
- 8.Kim, W.S., Steele, R.D., Ansar, A.I., Al, K., Nesnas, I.: Rover-Based visual target tracking validation and mission infusion. AIAA Space. 6717-6735 (2005)Google Scholar
- 11.Rumelhart, D., Hinton, E., Williams, R.: Learning internal representations by error propagating. In: Rumelhart, D., McClell, J. (eds.) Parallel Distributed Processing. MIT Press, Cambridge (1986)Google Scholar
- 12.Thompson, D.R., Castano, R.: Performance comparison of rock detection algorithms for autonomous planetary geology. In: Proc. of IEEE Aerospace Conf. paper no. 1251 (2007)Google Scholar