[1]张震,李丹丹..自适应双阈值的运动目标检测算法[J].郑州大学学报(工学版),2013,34(06):15-19.[doi:10.3969/j.issn.1671-6833.2013.06.004]
　ZHANGZhen,LI Dan-dan.An Adaptive DoubleThresholds Algorithm of Detecting Moving Objects[J].Journal of Zhengzhou University (Engineering Science),2013,34(06):15-19.[doi:10.3969/j.issn.1671-6833.2013.06.004]
点击复制

自适应双阈值的运动目标检测算法()
分享到：

《郑州大学学报(工学版)》[ISSN:1671-6833/CN:41-1339/T]

卷:
34卷
期数:
2013年06期
页码:
15-19
栏目:
出版日期:
2013-12-30

文章信息/Info

Title:
An Adaptive DoubleThresholds Algorithm of Detecting Moving Objects
作者:
张震; 李丹丹.
郑州大学电气工程学院,河南郑州,450001, 郑州大学电气工程学院,河南郑州,450001
Author(s):
ZHANGZhen; LI Dan-dan
School of Electrical Engineering，Zhengzhou University，Zhengzhou 450001，China
关键词:
双阈值;  运动掩膜;  函数链接型神经网络;  运动目标检测
Keywords:
double thresholds; motion object mask; functional chainneural network; moving object detection
分类号:
TP391.41
DOI:
10.3969/j.issn.1671-6833.2013.06.004
文献标志码:
A
摘要:
针对噪声、不同的天气状况和光照强度等环境变化对运动目标检测的影响,提出了一种自适应双阈值运动掩膜算法.为提高复杂环境条件下运动目标检测的识别率,该算法首先利用多帧平均法初始化背景,采用函数链接型神经网络算法动态更新高低两个阈值,自动适应光照变化.根据运动掩膜算法判定前景和背景区域动态更新背景后,采用自适应双阈值背景差法分割得到前景目标区域,并结合数学形态学方法,消除阴影,准确识别出前景目标.实验结果验证了该算法对运动目标检测的高准确性和良好的鲁棒性.
Abstract:
Due to the environmental change of the noise，different weather conditions andillumination，which influence the results of movingobject detection，thispaperproposed an adaptive double thresholds motionobiect mask algorithm．Toimprove the rate of motive vehicle recognition，this novel method first used multiple’ frame average algorithm to initialize the background，and adopted functional chain neural network method to uDdate thetwoof high andlow thresholds dynamically，which can adjust to changeable illumination automatic‘ 1y．According to the motion mask algorithm，the region of the foreground and background wasidentified and the currentbackgroundwasupdated．Then the region of the foreground object could be attained by dynamic double thresholds background difference method．Combined with the mathematical morphology method，much shadowwasdeleted and the foregroundobject wasrecognized correctly．The experimental results demonstrated that this detectingalgorithmwasmore accurate and robust．

参考文献/References:

[1]YANG Zhi-qi. A new algorithm of background imageextraction and update in the vehicle detection system[ C ]//2011 International Conference on MultimediaTechnology( ICMT )，Guangzhou，China:IEEEpress，2011 :5238 -5241.

[2]KIMA K，CHALIDABHONGEB T H，HARWOODAD，et al. Real-time foreground-background segmenta-tion using codebook model [J]. Real-Time Imaging，2005,3( 11):172 - 185.
[3]KORNPROBST P，DERICHE R，AUBERT G. Imagesequence analysis via partial difference equations [ J].Journal of Mathematical and Vision，1999，11 ( 1 ) :5- 26.

更新日期/Last Update: 1900-01-01