一种相关滤波中的模型自适应更新技术  

An Adaptive Model Updating Technology in Correlation Filtering

作　　者：江山 王录涛[1] 王奇 文成江 

机构地区：[1]成都信息工程大学计算机学院,四川 成都

出　　处：《计算机科学与应用》2022年第2期436-447,共12页Computer Science and Application

摘　　要：针对单一更新策略难以适应复杂多变的目标跟踪环境的问题,本文在ECO (高效卷积算子)算法基础之上,提出了一种基于旁瓣比的自适应策略更新跟踪算法。该算法通过对特征响应图中峰值旁瓣比的分析,实现了跟踪模型更新策略的自适应调整。同时使用OTB-100数据集对改进后的跟踪算法进行测试,经测试后的实验结果表明,相比原来的ECO算法,该算法可提升遮挡等复杂环境下目标跟踪的精度和鲁棒性。Aiming at the problem that a single update strategy is difficult to adapt to the complex and changeable target tracking environment, an adaptive strategy update tracking algorithm based on side lobe ratio is proposed based on ECO (efficient convolution operators) algorithm. By analyzing the peak to side lobe ratio in the characteristic response graph, the algorithm realizes the adaptive adjustment of the tracking model update strategy. At the same time, the OTB-100 data set is used to test the improved tracking algorithm. The experimental results show that compared with the original ECO algorithm, the algorithm can improve the accuracy and robustness of target tracking in complex environments such as occlusion.

关 键 词：相关滤波 目标跟踪 高效卷积算子 

分 类 号：TN713[电子电信—电路与系统]