Method of Remote Sensing Image Fine Classification Based on Geometric Features and SVM

Abstract:

In order to overcome the phenomenon of foreign bodies in the same spectrum in remote sensing images, as needs of land-use surveys, traditional methods often can’t get good results. In this paper, an efficient method for dividing the classifying results of traditional methods into further items is proposed and studied. Our approach to remote sensing image fine classification is based on both geometric features and SVM. By making full use of geometric features in the structure of property characteristics, we produce rules of the morphological characteristics and distribution. With deductive reasoning, we further classify the classifying results of SVM classifier. Tests showed that the method can be better broken down into the waters of rivers, lakes, reservoirs and ponds.

• Periodical: Key Engineering Materials (Volume 500)
• Pages: 562-568
• DOI: https://doi.org/10.4028/www.scientific.net/KEM.500.562
• Online since: January 2012
• Authors: Xiao Dong Zhou, Chun Cheng Yang, Ni Na Meng
• Keywords: Classification Rules, Fine Classification, Geometric Features, Image Classification, Support Vector Machine (SVM)

References

[1] Quinlan J R．Induction of Decision Trees [J]．Machine Learning, 1986, 1(1)：81-106.

[2] Quinlan J R．C4. 5：Programs for Machine Learning[M]．San Mateo, California：Morgan Kaufmann, 1993.

[3] Quinlan J R. Bagging, Boosting and C4. 5[DB/OL]. http: /citeseer. nj. nec. com/115849. html, (1996).

[4] Roli F, Fumera G. Support Vector Machines for Remote-sensing Image Classification [J]. Im age and Signal Processing for Remote Sensing, 2001, 4170: 160-166.

DOI: 10.1117/12.413892

[5] Luo J C, Zhou C H, Liang Y. Support Vector Machine for Spatial Feature Extraction and Classification of Remotely Sensed Imagery [J]. Journal of Remote Sensing, 2002, 6(1): 50-55.

[6] Huang Xueqiao and John R. Jensen. 1997. A Machine-Learning Approach to Automated Knowledge-Base Building for Remote Sensing Image Analysis with GIS Data. Photogrammetric Engineering & Remote Sensing, 63(10): 1185~1194.

[7] Vapnik, V.N., [The Nature of Statistical Learning Theory], Springer-Verlag, New York, (1995).

[8] Knerr, S., L. Personnaz, and G. Dreyfus (1990). Single-layer learning revisited: a stepwise procedure for building and training a neural network. In J. Fogelman (Ed. ), Neurocomputing: Algorithms, Architectures and Applications. Springer-Verlag.

DOI: 10.1007/978-3-642-76153-9_5

[9] Friedman, J. (1996). Another approach to polychotomous classification. Technical report, Department of Statistics, Stanford University. Available at http: /www-stat. stanford. edu/reports/friedman/poly. ps. Z.