Image segmentation is a technique used to segment region of interest (ROI) in the image for further image processing applications. Contour extraction is one of the most important image segmentation methods. In this paper, a contour extraction method is proposed, which consist of two stages. The first stage is automatic initial edge point selection (IEPS), and the second stage is segmental contour following (SCT) based on the edge points obtained at the first stage. Experiment result shows that, in comparing with other existing methods, more accurate contour can be obtained with low computational complexity using the proposed object contour extraction method, which will benefit to applications of industrial inspection or machine vision with lowlevel processor.