An active contour model is proposed for the problem of image segmentation. It has been applied to the tracking of deformable moving objects in the 2D image plane. For the edge detection of images, its energy function is defined and the object contour is recognized by minimization of the energy function. The active contour model is a useful method for tracking deformable objects, but it uses the energy function that is sensitive to the initialization of active contour. It is therefore hard to converge to a deep concave shape. Moreover, it is easy to detect a wrong contour by converging to the local minimum. Thus, in this paper, a new energy function which can overcome these weak points is proposed and it is applied to the tracking of moving objects. Experiments were carried out with the proposed method and the effectiveness of the new energy function was verified