计算机科学 ›› 2015, Vol. 42 ›› Issue (1): 312-316.doi: 10.11896/j.issn.1002-137X.2015.01.069

• 图形图像与模式识别 • 上一篇    

一种自适应的双门限场面运动目标检测方法

吴敏,吴宏刚,姚辉,王凯,蒋李   

  1. 中国民用航空总局第二研究所科研开发中心 成都610041,中国民用航空总局第二研究所科研开发中心 成都610041,中国民用航空总局第二研究所科研开发中心 成都610041,四川大学电气信息学院 成都610065,中国民用航空总局第二研究所科研开发中心 成都610041
  • 出版日期:2018-11-14 发布日期:2018-11-14
  • 基金资助:
    本文受机场增强视景监视系统(2013EG125056),机场(III级)综合交通监视与引导系统(2011BAH24B06)资助

Adaptive Double-threshold Detection Method of Airport Surface Moving Target

WU Min, WU Hong-gang, YAO Hui, WANG Kai and JIANG Li   

  • Online:2018-11-14 Published:2018-11-14

摘要: 为了有效解决在复杂环境下机场场面运动目标的精确检测问题,提出了一种自适应的双门限场面运动目标检测方法。首先采用混合高斯背景模型的方法来提取背景图像,然后使用两个门限值对差分图像进行前景目标分割,低门限阈值用于粗分割以检测出较明显的运动目标,在粗分割的基础上再用高门限阈值进行细分割以去除噪声目标和伪目标,最终得到场面运动目标的准确检测和分割结果。在复杂条件下的场景进行的实验,验证了该方法具有良好的噪声抑制能力和对慢目标良好的鲁棒性,同时能有效地分割出前景目标。

关键词: 双门限,混合高斯背景模型,背景图像,差分图像,前景目标

Abstract: To solve the accurate detection of air surface moving targets effectively under complicated environment,an adaptive double-threshold detection method of airport surface moving target was proposed.Firstly,Gaussian mixture model is adopted to extract the background image,and then foreground object of difference image is partitioned off by means of the two threshold value.Namely,low-threshold is used for coarse segmentation to detect moving object,which is obvious.In addition,high-threshold is used for fine segmentation to remove noise target and false target based on coarse segmentation.Finally,airport surface moving target is detected and segmented accurately.The experimental results of complicated background show that this method can avoid the noise,has robustness for the slower moving target and is effective for detecting.

Key words: Double-threshold,Gaussian mixture model,Background image,Difference image,Foreground object

[1] 胥欣,江登表,李勃,等.混合高斯模型运动检测算法优化[J].计算机应用研究,2013,30(7):2190-2194
[2] 林庆,徐柱,王士同,等.HSV自适应混合高斯模型的运动目标检测[J].计算机科学,2013,37(10):254-256
[3] 张帆,彭中伟,蒙水金.基于自适应阈值的改进Canny边缘检测方法[J].计算机应用,2012,2(8):2296-2298
[4] 康晓兵,魏生民.一种基于自适应阈值的图像伪造检测算法[J].计算机科学,2011,38(3):295-298
[5] Song Xue-hua,Chen Jing-zhu,Chong He,et al.A robust moving objects detection based on improved Gaussian mixture model [J].International Conference on Artificial Intelligence and Computational Intelligence,2010,2(6):54-58
[6] Gorur P,Amrutur B.Speed up Gaussian mixture model algo-rithm for background subtraction [C]∥IEEE International Conference on Advanced Video and Signal-based Surveillance.2011:386-391
[7] Maddalena L,Petrosino A.A self-organizing approach to background subtraction for visual surveillance applications [J].IEEE Transactions on Image Processing,2008,17(7):1168-1177
[8] 王正宁,刘昌忠,王娟,等.一种双门限场面运动目标检测系统[P].中国:201120089757.2011-11-16
[9] Xie Jin-quan,Chen Jin.An adaptive double-threshold spectrum sensing algorithm under noise uncertainty [J].IEEE International Conference on Computer and Information Technology,2012,35(9):824-827
[10] 高振华,黄山,张洪斌,等.基于自适应背景匹配的运动目标检测算法 [J].计算机工程与设计,2013,34(8):2839-2843
[11] 高美凤,刘娣.分块帧差和背景差相融合的运动目标检测 [J].计算机应用研究,2013,30(1):299-302

No related articles found!
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!