A visual mining based fame work for classification accuracy estimation

Classification techniques have been widely used in different remote sensing applications and correct classi-fication of mixed pixels is a tedious task. The problem is more complex with the classification of hyperspectral data and requires a thorough analysis. Traditional approaches adopt various statistical parameters, however does not facilitate effective visualisation. Data mining tools are proving very helpful in the classification process. We propose a visual mining based frame work for accuracy assessment of classification techniques using open source tools such as WEKA and PREFUSE. These tools in integration can provide an efficient approach for getting information about improvements in the classification accuracy and helps in refining training data set. We have illustrated frame work for investigating the effects of various resampling methods on classification accuracy and found that bilinear (BL) is best suited for preserving radiometric characteristics. We have also investigated the optimal number of folds required for effective analysis of LISS4 images.