An N-dimensional approach towards object based classification of remotely sensed imagery
This work addresses the need for improved land cover classification and urban analysis in remote sensing, though it appears incremental by combining existing techniques like support vector machines with object-based methods.
The paper tackled the problem of limited classification accuracy from pixel-based approaches in high-resolution remote sensing imagery by introducing an object-based classification scheme using a hierarchical support vector machine, which decreased class overlap and yielded higher classification accuracy and more accurate land cover maps, as evaluated in the Bhopal city study area.
Remote sensing techniques are widely used for land cover classification and urban analysis. The availability of high resolution remote sensing imagery limits the level of classification accuracy attainable from pixel-based approach. In this paper object-based classification scheme based on a hierarchical support vector machine is introduced. By combining spatial and spectral information, the amount of overlap between classes can be decreased; thereby yielding higher classification accuracy and more accurate land cover maps. We have adopted certain automatic approaches based on the advanced techniques as Cellular automata and Genetic Algorithm for kernel and tuning parameter selection. Performance evaluation of the proposed methodology in comparison with the existing approaches is performed with reference to the Bhopal city study area.