结合深度轮廓特征的改进孪生网络跟踪算法  

Improved Siamese network based object tracking combined with the deep contour feature

作　　者：余志超[1] 张瑞红 YU Zhichao;ZHANG Ruihong(School of Computer,Huanggang Normal University,Huanggang 438000,China)

机构地区：[1]黄冈师范学院计算机学院,湖北黄冈438000

出　　处：《西安电子科技大学学报》2020年第3期40-49,共10页Journal of Xidian University

基　　金：国家自然科学基金(21501061,71603098);湖北省自然科学基金(2018CFB597)。

摘　　要：针对现有的孪生网络目标跟踪算法存在跟踪漂移的问题,提出了一种结合深度轮廓生成网络的改进孪生网络跟踪模型,以实现复杂背景下对任何目标的稳定检测与跟踪。首先,轮廓检测网络自动获取目标的封闭轮廓信息,并利用泛洪聚类算法获得轮廓模板;然后将轮廓模板与搜索区域输入到改进的孪生网络,获得最优跟踪评分值,并自适应地更新轮廓模板。若目标被遮挡或跟踪丢失,则采用检测网络全视场搜索目标,实现全过程稳定跟踪。大量定性及定量仿真试验结果表明,这种改进模型不仅能够提高复杂背景下目标的跟踪性能,还能提升机载系统的反应时间,适合于工程应用。The existing Siamese object tracking algorithms easily lead to tracking drift under the influence of object deformation and occlusion,this paper proposes an improved object tracking algorithm based on deep contour extraction networks to achieve stable detection and tracking of any object under complex backgrounds.First,the contour detection network automatically obtains the closed contour information on the object and uses the flood-filling clustering algorithm to obtain the contour template.Then,the contour template and the search area are input into the improved Siamese network so as to obtain the optimal tracking score value and adaptively update the contour template.If the object is fully obscured or lost,the Yolov3 network is used to search the object in the entire field of view to achieve stable tracking throughout the process.A large number of qualitative and quantitative simulation results show that the improved model can not only improve the object tracking performance under complex backgrounds,but also improve the response time of airborne systems,which is suitable for engineering applications.

关 键 词：目标跟踪 深度学习 孪生网络 轮廓检测网络 目标检测 自适应模板更新 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程] TN219[自动化与计算机技术—控制科学与工程]