WEI Lin,CUI Rongyi,JIN Jingxuan*.Moving target detection algorithm based on gradientthreshold and feature suppression[J].Journal of Yanbian University,2017,43(01):37-42.
基于梯度阈值和特征抑制的运动目标检测算法
- Title:
- Moving target detection algorithm based on gradient threshold and feature suppression
- 分类号:
- TP391.41
- 文献标志码:
- A
- 摘要:
- 提出了一种基于梯度阈值和特征抑制的光流运动目标检测算法.首先将LK光流法和HS光流法思想进行互补,在梯度值较大的点使用亮度约束,梯度值较小的点使用平滑约束; 然后采用特征抑制方法,筛选能够确定运动目标位置的光流点; 最后检测出完整的运动目标.实验结果表明,本文算法提高了运动目标检测的准确度,而且实时性较好.
- Abstract:
- In this paper, we propose a new optical flow moving object detection algorithm based on gradient threshold and feature suppression. First we adopt LK optical flow method and HS optical flow method to complement each other. We use the brightness constraints on the points with large gradient and the smooth constraints on those with small gradient. Then combining with feature suppression method, the algorithm screen the optical flow points to determine the location of moving, and ultimately moving target. Experimental results show that the proposed method can improve the detection accuracy of real moving objects, and has better real-time performance.
参考文献/References:
[1] 华媛蕾,刘万军.改进混合高斯模型的运动目标检测算法[J].计算机应用,2014,34(2):580-584.
[2] 袁国武,陈志强,龚健,等.一种结合光流法与三帧差分法的运动目标检测算法[J].小型微型计算机系统,2013,34(3):669-671.
[3] 杨叶梅.基于改进光流法的运动目标检测[J].计算机与数字工程,2011,263(9):108-110.
[4] 高磊.基于光流的动态场景中运动车辆检测与跟踪算法研究[D].合肥:中国科学技术大学,2014.
[5] 刘洪彬.权重系数自适应光流法运动目标检测[J].光学精密工程,2016,24(2):460-468.
[6] 安博.动态背景下运动目标检测的研究[D].天津:天津理工大学,2008.
[7] Verri A, Poggio T. Against quantitative optical flow[C]// Proceedings of the First International Conference on Computer Vision. Washington: IEEE Comput Soc Press, 1987:171-180.
[8] 卫保国,李晶.一种针对大尺度运动的快速光流算法[J].计算机应用研究,2012,29(9):3551-3557.
[9] 黄杭.基于光流法的运动目标检测与跟踪[D].沈阳:东北大学,2014.
[10] CHEN Junchao, ZHANG Junhao, LIU Shijia, et al. Improved target detection algorithm based on background modeling and frame difference[J]. Computer Engineering, 2011,37:171-173.
[11] 王振亚,曾黄麟.一种基于帧间差分和光流技术结合的运动车辆检测和跟踪新算法[J].计算机应用与软件,2012,29(5):117-120.
[12] 王天召,徐克虎,黄大山.动态背景下的运动目标检测[J].计算机技术与发展,2013,23(7):104-107.
[13] Horn B K P, Schunck B G. Determining optical flow[J]. Artificial Intelligence, 1981,17(1):185-203.
备注/Memo
收稿日期: 2016-12-06
*通信作者: 金王景璇(1972—),女,副教授,研究方向为智能信息处理.