The new generation of Very High Resolution Satellite Imagery (VHRSI) offers a mapping potential for large scale maps. Many features like buildings, roads and green areas could be extracted. Manual techniques are fading away as they are inefficient and time consuming. Thus, the solution is increasing the automation process in order to improve the efficiency of satellite topographic mapping. This research tries to set up a work follow for automatic feature extraction from VHRSI. Fifteen classification techniques were applied. The one meter pan sharpened IKONOS imagery are used to extract features that were compared against the already exist 1/5,000 maps. Two experiments were conducted. For the first case, the classification is carried using the satellite images only as an input for the classification process. While for the second case, the classification is carried using the satellite imagery plus a Digital Surface Model (DSM) as an additional layer for the classification process. The classified images are then processed through a series of image processing process to produce the digital vector map. For both cases all results are tabulated, analyzed and recommendations are mentioned.