(An) active contour model using image flow for tracking a moving object

Abstract

Active contour models were developed for useful segmenting and tracking tools for rigid or non-rigid (i.e. deformable) objects and Snake, one of the Active contour models (ACMs), was introduced by Kass in 1987. Snake is desiged on the basis of Snake energies and segmenting and tracking can be executed successfully by the process of energy minimization. The contraction is a important process for segmenting objects from images, but the contraction forces of Kass`` Snake are dependent on the object``s forms. In this research new contracion energy independent of the object``s forms was proposed for the segmentation. Kass`` Snake can be applicable to the case of small changes between two images when its solutions can be achieved on the basis of variational approach. In this research the image flow energy was newly proposed to make Snake``s nodes roll along the direction of motion by using the information of image flow in this large moving case. Furthermore considering the case that the motion is too large to apply the the image flow energy for tracking, the jump mode was proposed for solving the problem of large change between two iamges in this research. The vector used to make Snake``s nodes jump to the new location can be gained by processing of image flow. The effectiveness of the proposed Snake was confirmed by some simulations on the tracking a object.