A comparative study of ANN, k-means and AdaBoost algorithms for image classification

Data mining is the method of extracting the valuable systematic information from huge databases. Image classification has constantly been a vital task for several applications such as remote sensing medical field, pattern recognition. It converses to the task of removing information classes from a multiband raster image. The resolving of the classification method is to classify all pixels in a one image class into another class. The target of image classification is to find the exclusive dark level of images. This paper concentrates on the study of artificial neural network, Adaboost and k-means algorithms in image classification.