Image segmentation and boundary extraction are very important in the fields of image understanding, image analysis, pattern recognition and computer vision etc., and geometric active contour model (GACM) is one of the most important tools in the areas of image segmentation and boundary extraction. While before using this model initial contour must be selected. In order to realize initial contour automatic selection of geometric active contour model, marker-based watershed algorithm is adopted to extract object contour in the paper, and then the extracted object contour is taken as initial contour of geometric active contour model for precise segmentation. The result from comparative experiments shows that using watershed algorithm not only can realize the automatic selection of initial contour, reduce manual intervention, but also can reduce iteration numbers of curve evolution, improve segmentation efficiency of the model. It also can segment images of complex background efficiently.