The novel algorithm for motion estimation and prediction of objects in dynamic scene by processing a limited sequence of images is presented in this paper. The proposed hybrid algorithm is based on computing differences between consecutive frames of video streams for fast detection of objects in motion by processing of their time-varying edges in real time applications. For quantitative prediction of motion and computing of its characteristics, the interpolation technique has been used in combination with the motion stabilization technique useful for analysis of irregular displacement of objects. The principal goal of proposed algorithm is the design of simple but efficient facilities for motion prediction when the possible routes may be fast computed even with occlusions and rotation. The introduced algorithm has been tested in order to evaluate its utility and performance.