电子学报 ›› 2022, Vol. 50 ›› Issue (7): 1558-1566.DOI: 10.12263/DZXB.20210641

• 学术论文 • 上一篇    下一篇

基于前景优化的视觉目标跟踪算法

谢青松1, 刘晓庆2, 安志勇1, 李博1   

  1. 1.山东工商学院计算机科学与技术学院,山东 烟台 264000
    2.山东工商学院信息与电子工程学院,山东 烟台 264000
  • 收稿日期:2021-05-19 修回日期:2021-10-07 出版日期:2022-07-25
    • 作者简介:
    • 谢青松 男,1965年1月生于山东青岛,现为山东工商学院计算机科学与技术学院教授、硕士生导师,主要研究方向为智能算法、图像识别、目标跟踪.E-mail: qs_xie@163.com
      刘晓庆 女,1996年5月生于山东青岛,现为山东工商学院信息与电子工程学院硕士研究生,主要研究方向为目标跟踪.E-mail: 1322699199@qq.com
      安志勇(通讯作者) 男,1975年10月生于山西怀仁,工学博士,现为山东工商学院计算机科学与技术学院副教授,研究方向为计算机视觉、目标跟踪等.E-mail:azytyut@163.com
      李 博 男,1980年3月生于辽宁锦州,东北大学信息科学与工程学院计算机系统结构博士,研究领域为大数据与机器学习.
    • 基金资助:
    • 国家自然科学基金 (62072285); 山东省自然科学基金 (ZR202102230438); 山东省重点研发计划 (软科学) (2020RKB01017); 山东工商学院校级教学改革项目 (11688202023)

Visual Object Tracking Algorithm Based on Foreground Optimization

XIE Qing-song1, LIU Xiao-qing2, AN Zhi-yong1, LI Bo1   

  1. 1.School of Computer Science and Technology, Shandong Technology and Business University, Yantai, Shandong 264000, China
    2.School of Information and Electronic Engineering, Shandong Technology and Business University, Yantai, Shandong 264000, China
  • Received:2021-05-19 Revised:2021-10-07 Online:2022-07-25 Published:2022-07-30
    • Supported by:
    • National Natural Science Foundation of China (62072285); Natural Science Foundation of Shandong Province, China (ZR202102230438); Key Research and Development Project of Shandong Province (Soft Science) (2020RKB01017); Teaching Reform Project of Shandong Technology and Business University (11688202023)

摘要:

将目标分割技术引入跟踪领域是当前的研究热点.目前,基于分割的跟踪算法往往根据分割结果计算最小外接矩形,以此作为跟踪框,但复杂的目标运动使得跟踪框内包含较多背景,从而导致精度下降.针对该问题,本文提出了一种基于前景优化的视觉目标跟踪算法,将跟踪框的尺度和角度优化统一于前景优化框架中.首先评估跟踪框内的前景比例,若小于设定阈值,则对跟踪框分别进行尺度和角度优化;在尺度优化模块中,结合回归框计算跟踪框的条件概率,根据条件概率的结果分情形进行尺度优化;角度优化模块中,针对跟踪框设定多个偏移角度,利用前景IoU(Intersection over Union)极大策略选择最优跟踪框角度.结果证明,将本文方法应用于SiamMask算法,精度在VOT2016,VOT2018和VOT2019数据集分别提升约3.2%,3.7%和3.6%,而EAO分别提升约1.8%,1.9%和1.6%.另外,本文的方法针对基于分割的跟踪算法具有一定的普适性.

关键词: 目标分割, 目标跟踪, 前景优化, 尺度优化, 角度优化

Abstract:

The introduction of object segmentation technology into the tracking field is a current research hotspot. At present, the tracking algorithm based on segmentation often calculates the minimum bounding rectangle as the bounding box according to the segmentation result. However, the complex target movement makes the bounding box contain more background, which leads to a decrease in accuracy. In response to the problem, this paper proposes a visual object tracking algorithm based on foreground optimization, which unifies the optimization of the scale and angle in the bounding box into the foreground optimization frame. First, the foreground ratio in the bounding box is evaluated. If it is less than the set threshold, the scale and angle of the bounding box are optimized; in the scale optimization module, the conditional probability of the bounding box is calculated in combination with the regression box, and the scale is optimized according to the results of the conditional probability; in the angle optimization module, many deviation angles are set for the bounding box, and the optimal bounding box angle is chosen by the foreground IoU (Intersection over Union) maximum strategy. The proposed method is applied to the SiamMask algorithm. Results show that the accuracy is improved by about 3.2%, 3.7% and 3.6% in the VOT2016, VOT2018 and VOT2019 data sets, respectively, while EAO is increased by about 1.8%,1.9% and 1.6%, respectively. Moreover, our method has a certain universality for segmentation-based tracking algorithms.

Key words: object segmentation, object tracking, foreground optimization, scale optimization, angle optimization

中图分类号: