视频监控领域基于多特征融合与自适应模型更新的目标跟踪

Object Tracking Algorithm Based on Adaptive Updating and Multi-feature Fusion in Intelligent Video Surveillance

摘要:

随着视频处理技术和网络技术的发展,视频监控应用逐渐渗透到了人们日常活动中的方方面面,如何设计实现精度高、鲁棒性好的目标跟踪技术仍然是当今研究的热点及难点。本文在工程应用实践的基础上,提出一多特征融合与自适应模型更新的空时上下文目标跟踪算法,通过将丰富多样的多特征信息整合到空时上下文模型中。由于多特征具有互补特性,可以克服单一特征对目标区域描述不足的缺陷,提升算法的抗干扰能力。同时,本文也提出了一种自适应学习因子策略,增强了模型的泛化能力。大量的仿真实验结果表明本文所提算法的跟踪性能超过传统的空时上下文文目标跟踪算法,对复杂的跟踪场景具有更强的鲁棒性与抗干扰能力。

Abstract:

#$NL: With the development of video processing and network technology, video surveillance applications gradually penetrated into every aspect of people"s daily activities.How to design a object tracking technique with high precision and robustness is still a hotspot and difficulty in current research. This paper proposes an improved spatio-temporal contexttracking algorithm based on multi-feature fusion and adaptive model updating. Based on the spatio-temporacontext tracking algorithm, our proposed algorithm integrates multi-feature informations into the spatio-temporacontextmodel. Since the complementary characteristics of multiple features, it is possible to overcome the disadvantages of the single feature and improve the anti-jamming ability. In addition, this paper also proposes an adaptive learning factor strategy to enhance the generalization ability of the model. A large number of simulation results show that the tracking performance of our proposed algorithm outperforms the traditional spatio-tempora contexttracking algorithm, and has stronger robustness and anti-jamming capability for complex scenes.#$NLKeywords:Object tracking; Multi-feature fusion; Adaptive; Spatio-tempora; Generalization ability; Complementary property