一种动态场景下运动对象分割新算法

J4 ›› 2012, Vol. 34 ›› Issue (4): 43-46.

一种动态场景下运动对象分割新算法

马志强，张晓燕，朱子健，张锐

空军工程大学电讯工程学院网络工程系，陕西西安 710077）

收稿日期:2011-11-05 修回日期:2012-02-10 出版日期:2012-04-26 发布日期:2012-04-25
基金资助:
博士后科学基金特别资助项目（201104787）;博士后科学基金资助项目（20100471838）;陕西省自然基金资助项目（2010JM8014）

A Novel Approach for Moving Object Segmentation Used in Dynamic Scenes

MA Zhiqiang，ZHANG Xiaoyan，ZHU Zijian，ZHANG Rui

（Department of Network Engineering,School of Telecommunication Engineering,Air Force Engineering University,Xi’an 710077,China）

Received:2011-11-05 Revised:2012-02-10 Online:2012-04-26 Published:2012-04-25

摘要/Abstract

摘要：

视频运动对象分割是计算机视觉和视频处理的基本问题。在摄像机存在全局运动的动态场景下，准确分割运动对象依然是难点和热点问题。本文提出一种基于全局运动补偿和核密度检测的动态场景下视频运动对象分割算法。首先，提出匹配加权的全局运动估计补偿算法，消除动态场景下背景运动对运动对象分割的影响；其次，采用非参数核密度估计方法分别估计各像素属于前景与背景的概率密度，通过比较属于前景和属于背景的概率及形态学处理得到运动对象分割结果。实验结果证明，该方法实现简单，有效地提高了动态场景下运动对象分割的准确性。

关键词: 动态场景, 运动对象, 运动补偿, 核密度检测

Abstract:

Video moving object segmentation is a basic problem of computer vision and video processing. In the video sequences of a dynamic scene which has global (camera) motion, accurate moving object segmentation is still a key and hot research topic. A novel video moving object segmentation algorithm based on global motion compensation and nonparametric kernel density estimation is proposed in this paper. Firstly, an efficient and accurate global motion compensation method is used to remove the motion of the background. Then the nonparametric kernel density estimation is applied to establish foreground/background probability models. Finally, the moving object can be obtained by comparing the foreground/background probability and morphological postprocessing. The experimental results demonstrate that the proposed algorithm has good results and reduces the complexity of moving object segmentation in dynamic scenes.

Key words: dynamic scene;moving object;motion compensation;kernel density estimation

马志强，张晓燕，朱子健，张锐. 一种动态场景下运动对象分割新算法[J]. J4, 2012, 34(4): 43-46.

MA Zhiqiang，ZHANG Xiaoyan，ZHU Zijian，ZHANG Rui. A Novel Approach for Moving Object Segmentation Used in Dynamic Scenes[J]. J4, 2012, 34(4): 43-46.

[1]	戴康佳, 徐慧英, 朱信忠, 黄晓, 李琛, 刘巍, 曹雨淇, 王拔龙, 刘子洋, 陈国强. 基于轻量化目标检测网络的RGB-D视觉SLAM系统[J]. 计算机工程与科学, 2024, 46(11): 2017-2026.
[2]	秦龙龙1,钱渊1,侯雪2，张晓燕1. 基于维纳线性预测的Horn-Schunck光流运动矢量优化算法[J]. J4, 2015, 37(07): 1360-1365.
[3]	王亮芬,周东翔,梁华,蔡宣平. 动摄像机下基于SIFT特征匹配和MHI的目标检测[J]. J4, 2010, 32(1): 67-70.
[4]	倪崇嘉[1] 刘文奇[2] 张爱英[2]. 基于数学形态学的视频图像序列中的运动目标检测[J]. J4, 2006, 28(6): 69-70.
[5]	赵彦玲[1] 张之超[2] 高振明[2] 张庆超[3]. 图像序列中运动对象的自动提取方法[J]. J4, 2006, 28(2): 63-65.
[6]	黄金权王程王润生. 基于投影特征的快速图像稳定算法[J]. J4, 2006, 28(1): 82-85.
[7]	余胜生仲玲周敬利陈加忠. 基于H．26L的运动估计算法的一种改进[J]. J4, 2003, 25(6): 1-2.