Classification of satellite Images is one of the major research areas in remote sensing fields. Classification of remote sensed data is required for accurate classification of semi urban land features. Satellite image classification plays an essential role in proper monitoring and management of natural and manmade resources on the earth surface. However a good data set is required for the accurate classification of remotely sensed data. In this paper, to classify the data set, various image fusion techniques are used for fusing high resolution Panchromatic data with low resolution Multi-spectral data which gives better quality and more informative image data set. The performances of different fusion techniques are then evaluated to identify the best possible technique which gives better result for image classification.