Moving Object Tracking Method Using Improved Camshift With Surf Algorithm
Abstract— Object detection and tracking has become one of the most active research areas in the past few years. Many
detection and tracking techniques are based on color and feature point matching. However, these techniques are characterized
by either high computational complexity or reduced detection quality. We propose two different algorithms for real time
dynamic target tracking, namely Improved Continuously Adaptive Meanshift (CAMshift) and Speeded Up Robust Features
(SURF). We also propose a method to judge whether the object is lost. CAMshift uses color features and is sensitive to some
environmental factors. When similar colors are existing in the background, traditional CAMshift algorithm may fail, that is
the target getting lost. To solve this problem, an improved CAMshift algorithm is firstly proposed in this paper to reduce the
influence of illumination interference. Bhattacharya distance method is proposed to judge whether the target is lost. Once the
target is judged lost, the SUR method is utilized to find it again and the improved CAMshift method keeps on tracking the
target continuously. SURF is invariant to scale, rotation and translation of images. We have developed a Human Computer
Interaction (HCI) application for handling basic Windows OS operations using gesture input based on the proposed method.
We program in C++ based on OpenCV. The results prove that the proposed method is more robust than the traditional
CAMshift and give better tracking performance than some other improved methods.