Classification of a quickbird satellite image by Machine learning techniques: Mapping an urban Environement by decision tree method

Classification is a crucial stage in the processing of satellite images that influence considerably quality result. A variety methods proposed literature for purposes image classification. They present many differences their basic principles, thus results obtained. Therefore, study different classification seems to be essential. The with conventional can done several ways using algorithms. These algorithms divided into two main categories: supervised and non-supervised. Decision tree on contrary machine learning tool. It plain model characterized by simplicity understanding interpretation. This work aims firstly, classify high resolution Quickbird an urban area decision method compare it order evaluate its efficiency. methodology consists stages: evaluation results. second based calculation number statistical indices derived from confusion matrix: parameter “kappa’ overall coefficient precision.