Rough Sets: Theoretical Aspects of Reasoning about Data
Rough Sets: Theoretical Aspects of Reasoning about Data
Granular Approach to Object-Oriented Remote Sensing Image Classification
RSKT '09 Proceedings of the 4th International Conference on Rough Sets and Knowledge Technology
Hi-index | 0.00 |
This paper proposes a method to solve the selection problem in GIS generalization by leveraging the rough sets theory for attribute reduction. In specific, by taking into account the special characteristics of the GIS spatial data, our method can be outlined as follows. First, we discretize the continuous-valued attributes through unsupervised discretization method; Second, we classify in a fuzzy manner the spatial objects, whose result will then serve as the decisional attributes; Third, we evaluate the respective importance of these attributes through the attribute reduction method borrowed from the rough sets theory and consequently we conduct a dynamic sorting according to the resulting importance values. Through experimentation results, the effectiveness performance of our proposed method is validated.