A Template Update Cam Shift Algorithm Based on LTP Texture

Abstract

In view of the illumination changes and the deformations of objects in the traffic environment, a template update object tracking algorithm which is based on the fusion of color and texture features is proposed. Under the framework of traditional algorithm, the algorithm not only takes the weighted color histogram to describe the object's color, but also fuses LTP texture operator. Finally, the object was described by the weighted fusion of both and the template was set up to update the weights of different templates. The contrast tracking experiments were carried out under the environment of some illumination changes and the deformations of objects. And then the robustness of the algorithm was verified on the premise of the real-time performance.