Images with multi-spectral bands give rich data for features witch is useful in description, classification and extraction. The research objective is to investigate the optimum bands for classification from high resolution multispectral images. World View3 (WV3) satellite provides eight spectral bands in addition to panchromatic image with high resolution spatial information that give details of feature of captured area. Image classification is applied using Maximum Likelihood (ML) technique using a combination of multi-spectral bands of WV3 image. Spectral bands are used to form sets of input data as cases of study. The classification is applied using all spectral bands after applying common atmospheric correction procedure of generating reflectance image of the captured area. The results of classification of each case of study are assessed versus a classification result using reflectance image with same classification technique. The research work flow is processed by using ENVI and MATLAB environments. The assessment of classification results is represented in confusion matrix format and determination of Kappa coefficients. The preliminary results show that a set of (Near Infrared (NIR), Green (G) and Costal (C) bands) gives accurate classification result for the area of study with overall accuracy 97.9629% and it is evaluated by Kappa coefficient which is 0.9460