Consensual and Hierarchical Classification of Remotely Sensed Multispectral Images

Consensual and hierarchical approaches are applied in classification of remotely sensed multispectral images. The proposed method consists of nonlinear image filters, three different types of classifiers which use hierarchical neural networks with rejection schemes, and a combining scheme by a consensus rule. By nonlinear image filtering, class separability is improved. By successive classifiers which are tuned to reduce remaining error, classification performance increases. This structure includes detection schemes to decide whether successive classifiers are utilized for each input. The classification results by multiple classifiers using hierarchical neural networks are combined by a consensus rule to obtain more reliable and accurate results based on group decision. Keywords-consensus; classification; neural networks; rejection schemes; hierarchical