[1]刘志峰,陈姚节,程 杰.结合帧差法的尺度自适应核相关滤波跟踪[J].计算机技术与发展,2021,31(02):70-74.[doi:10. 3969 / j. issn. 1673-629X. 2021. 02. 013]
　LIU Zhi-feng,CHEN Yao-jie,CHENG Jie.Scale Adaptive Kernel Correlation Filter Tracking Combined with Frame Difference Method[J].,2021,31(02):70-74.[doi:10. 3969 / j. issn. 1673-629X. 2021. 02. 013]
点击复制

结合帧差法的尺度自适应核相关滤波跟踪()
分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
31
期数:
2021年02期
页码:
70-74
栏目:
图形与图像
出版日期:
2021-02-10

文章信息/Info

Title:
Scale Adaptive Kernel Correlation Filter Tracking Combined with Frame Difference Method
文章编号:
1673-629X(2021)02-0070-05
作者:
刘志峰1; 陈姚节1; 2; 3; 程 杰1
1. 武汉科技大学 计算机科学与技术学院,湖北 武汉 430065;?
2. 智能信息处理与实时工业系统湖北省重点实验室,湖北 武汉 430065;?
3. 冶金工业过程国家级虚拟仿真实验教学中心,湖北 武汉 430065
Author(s):
LIU Zhi-feng1; CHEN Yao-jie1; 2; 3; CHENG Jie1
1. Department of Computer Science and Technology,Wuhan University of Science and Technology,Wuhan 430065,China;?
2. Hubei Province Key Laboratory of Intelligent Information Processing and Real-time Industrial System,Wuhan 430065,China;?
3.?Metallurgical Industry Process National Virtual Simulation Experimental Teaching Center, Wuhan 430065,China
关键词:
目标跟踪; 相关滤波; 帧差法; 尺度自适应; 模型更新
Keywords:
object tracking; correlation filter; frame difference method; scale adaptive; model update
分类号:
TP391
DOI:
10. 3969 / j. issn. 1673-629X. 2021. 02. 013
摘要:
为处理复杂应用场景下核相关滤波器跟踪效果不理想的问题,提出了一种结合帧差法的尺度自适应核相关滤波跟踪算法。 在训练得到相关滤波器后,借助帧差法来处理下一帧图像,获得目标的预测位置,扩充算法的检测区域;然后通过尺度池构建多尺度待检测图像块集,通过相关滤波器来求得最大响应,估计出目标的最佳位置和最佳尺度; 最后利用平均峰值相关能量(average peak-to correlation energy, APCE)作为跟踪置信度指标,引入高置信度更新机制,在目标被遮挡时,停止更新模型,防止误差被积累,提高正确率。 在 OTB100 数据集上与若干视觉跟踪算法进行了对比实验,改进算法的成功率和距离精度均表现最优,比 KCF 算法高出 21.7 个百分点和 12.0 个百分点。 该算法在目标快速运动、尺度变化、遮挡等复杂场景下,均具有较强的精确性和鲁棒性。
Abstract:
To deal with the problem of unsatisfactory tracking effect of kernel correlation filter under complex application scenarios,a scale adaptive kernel correlation filter tracking algorithm combined with frame difference method is proposed. After training the correlation filter,by means of the frame difference method,the next frame of image is processed,the predicted position of the target is obtained, and the detection area of the algorithm is expanded. Then,the multi-scale set of image blocks is built through the scale pool,and the maximum response is obtained through the correlation filter,and the optimal position and scale of the target are estimated. Finally, the average peak-to correlation energy (APCE) is used as the tracking confidence index, and a high confidence update mechanism is introduced. When the target is occluded,the updating model is stopped to prevent errors from being accumulated and improve accuracy.Compared with several visual tracking algorithms on the OTB100 database, experiments are performed on the improved algorithm,and the success rate and distance accuracy of the improved algorithm are the best,which are 21.7% and 12.0% higher than the KCF algorithm.The proposed algorithm has high accuracy and strong robustness in complex scenes such as rapid target movement,scale change,and occlusion.

相似文献/References:

[1]刘翔 吴谨 祝愿博 康晓晶.基于视频序列的目标检测与跟踪技术研究[J].计算机技术与发展,2009,(11):179.
　LIU Xiang,WU Jin,ZHU Yuan-bo,et al.A Study of Object Detecting and Tracking Based on Video Sequences[J].,2009,(02):179.
[2]雷云 王夏黎 孙华.基于视频的交通目标跟踪方法研究[J].计算机技术与发展,2010,(07):44.
　LEI Yun,WANG Xia-li,SUN Hua.The Research about Transport Target Tracking Based on Video[J].,2010,(02):44.
[3]谢之宇 蒋晓瑜 汪熙 裴闯.基于多线索融合的目标跟踪算法研究[J].计算机技术与发展,2011,(03):125.
　XIE Zhi-yu,JIANG Xiao-yu,WANG Xi,et al.A Target Tracking Algorithm Research Based on Multi-Cue Fusion[J].,2011,(02):125.
[4]尤天来 周海徽[].红外目标跟踪技术研究[J].计算机技术与发展,2011,(10):140.
　YOU Tian-lai,ZHOU Hai-hui.Research of Infrared Target Tracking Technology[J].,2011,(02):140.
[5]赵侃 漆德宁.基于UKF滤波的FDOA和TDOA联合定位跟踪算法[J].计算机技术与发展,2012,(05):127.
　ZHAO Kan,QI De-ning.A Tracking TDOA/FDOA Joint Location Algorithm Based on UKF[J].,2012,(02):127.
[6]姚放吾 许辰铭.基于目标质心的Meanshift跟踪算法[J].计算机技术与发展,2012,(06):104.
　YAO Fang-wu,XU Chen-ming.A Meanshift Tracking Algorithm Based on Centroid[J].,2012,(02):104.
[7]张璐,张国良,张维平,等.改进IMM算法在机器人目标跟踪中的应用[J].计算机技术与发展,2013,(02):149.
　ZHANG Lu,ZHANG Guo-liang,ZHANG Wei-ping,et al.Application of Improved IMM Algorithm in Robot Target Tracking[J].,2013,(02):149.
[8]吴佳家,高珏,李敏,等.基于均值漂移和卡尔曼滤波的跟踪算法研究[J].计算机技术与发展,2014,24(01):5.
　WU Jia-jia[,GAO Jue[],LI Min[],et al.Research on Target Tracking Algorithm Based on Mean Shift and Kalman Filter[J].,2014,24(02):5.
[9]郝杰,任静. 高超声速飞行器交互式多模型跟踪算法仿真[J].计算机技术与发展,2015,25(02):204.
　HAO Jie,REN Jing. Tracking Algorithm Simulation of Interactive Multiple Model for Hypersonic Flight Vehicle[J].,2015,25(02):204.
[10]高翔,朱婷婷,刘洋. 多摄像头系统的目标检测与跟踪方法研究[J].计算机技术与发展,2015,25(07):221.
　GAO Xiang,ZHU Ting-ting,LIU Yang. Research of Target Detection and Tracking Method for Multi-camera System[J].,2015,25(02):221.
[11]马媛媛,杨小军.基于多层深度特征的尺度相关粒子滤波算法[J].计算机技术与发展,2021,31(06):169.[doi:10. 3969 / j. issn. 1673-629X. 2021. 06. 030]
　MA Yuan-yuan,YANG Xiao-jun.A Scale Dependent Particle Filter Algorithm Based onMulti-layer Depth Characteristics[J].,2021,31(02):169.[doi:10. 3969 / j. issn. 1673-629X. 2021. 06. 030]

更新日期/Last Update: 2020-02-10