Research On Complex Condition Of The Object Tracking Technology

Visual tracking is a new cutting-edge technology in recent years, which involves technologies of computer vision, pattern recognition, artificial intelligence, automatic control system and so forth. A lot of potential applications have been used in our real life and performance a good promising. However, Visual tracking also has lots of challenge to been done, such as multi-scale, partial occlusion or most occlusion, light change and noise which interfering the object tracking. Although many effective visual tracking systems have been proposed to solve the tracking robust, tracking accuracy and real-time, the traditional tracking methods also have other challenges.This paper focuses on the visual tracking in complex condition based on the traditional tracking algorithm. Two new groups of Haar-like feature were proposed in this article and achieved better performance both in highly detection accuracy and less weak classifiers. Then, we improves the TLD algorithm and proposes local and global search based on Sliding-window methods, Integral Histogram Filter and Random Haar-like Feature Filter to solve the easily drift problem of the traditional tracking algorithm in complex conditions. First, we exploit the Integral Histogram Filter to reject the Sliding-window patches as quickly as possible to release the feature matching in the following filters. Then, we use Random Haar-like Feature Filter to overcome drift problem which causes the less accuracy during the object tracking in complex conditions (multi-object, partial or most occlusion, fast movement). We ultimately combine filters of the TLD algorithm and two new filters of our proposed. The experimental results show that the proposed approaches compared with the traditional tracking algorithms not only presents robust and tracking accuracy in stable background or complex conditions, but also obtains the best computing speed with the use of the local and global search. The proposed method is able to detect the multi-scale object accurately both in different environment and tracking object deformation.