基于高斯混合模型和canny算法的运动目标检测

广东工业大学学报 ›› 2011, Vol. 28 ›› Issue (3): 87-91.

基于高斯混合模型和canny算法的运动目标检测

1.嘉应学院计算机学院，广东梅州 514015；2.广东工业大学信息工程学院，广东广州 510006；3.仲恺农业工程学院信息学院，广东广州

出版日期:2010-10-06 发布日期:2010-10-06
作者简介:陈世文（1983-），男，实验员，硕士，主要研究方向为目标检测、目标跟踪等.
基金资助:
国家自然科学基金资助项目（61001179）；广东省自然科学基金资助项目（07301038, 9451009001002667）

Detection of Moving Objects Based on the Gaussian Mixture Model and the Canny Operator

1.Department of Computer Science, Jiaying University, Meizhou 514000, China； 
2.Faculty of Information Engineering, Guangdong University of Technology, Guangzhou 510006, China；
3.Faculty of Information, Zhongkai University of Agriculture and Engineering, Guangzhou 510225, China

Online:2010-10-06 Published:2010-10-06

摘要/Abstract

摘要： 提出一种基于高斯混合模型和canny算法的运动目标检测算法.利用高斯混合模型计算像素之间的颜色信息，同时利用高斯混合模型更新背景信息；用canny算子提取图像的边缘信息；将颜色信息和区域结构信息线性融合起来，较好地解决了边缘信息明显的运动目标检测.实验中采用改进的加权高斯模型及传统的canny算法相结合.结果表明，本文方法比经典高斯混合模型方法具有较高的分割精度，鲁棒性较好.

关键词: 高斯混合模型；目标检测；canny算子

Abstract: A method of detecting moving objects is proposed by combining the Gaussian mixture model with the canny operator. First, the Gaussian mixture model was employed to extract color information and to update the background information. Second, the canny operator was employed to extract edge information. Last, color information and part of area information were integrated for segmentation, which improved the ability to detect moving objects where edge information was distinct. An improved Gaussian mixture model and a traditional canny algorithm were employed in the experiment. Experimental results indicate that the proposed method is superior to the traditional Gaussian mixture model.

Key words: Gaussian mixture model, object detection, canny operator

陈世文1, 2, 蔡念2, 肖明明3. 基于高斯混合模型和canny算法的运动目标检测[J]. 广东工业大学学报, 2011, 28(3): 87-91.

CHEN Shi-Wen1, 2 , Cai-Nian2, Xiao-Ming-Ming3. Detection of Moving Objects Based on the Gaussian Mixture Model and the Canny Operator[J]. Journal of Guangdong University of Technology, 2011, 28(3): 87-91.

参考文献

［1］ Suo Peng，Wang Yanjiang. An improved ddaptive background modeling algorithm based on gaussian mixture model［C］.ICSP2008, 1426-1439

［2］ Power P W, Schoonees J A. Understanding background mixture models for foregrounds segmentation ［C］∥ Proceedings of Image and Vision Computing. New Zealand: Auckland, 2002: 267-271.

［3］陈世文，蔡念，唐孝艳. 一种基于高斯混合模型的运动目标检测的改进算法［J］. 现代电子技术, 2010, 33 (2): 125127,130.

［4］ Harville M, Gordon G, Woodfill J. Foreground segmentation using adaptive mixture models in color and depth ［C］∥IEEE Proceedings of IEEE Workshop on Detection and Recognition of Events in Video. Canada Vancouver, 2001: 3-11.

［5］原春锋，王传旭，张祥光,等. 光照突变环境下基于高斯混合模型和梯度信息的视频分割［J］ .中国图形图像学报，2007，11（12）：2 068-2 073.

［6］刘鑫，刘辉，强振平,等.混合高斯模型和帧间差分相融合的自适应背景模型［J］ .中国图形图像学报，2008，4（13）：729-735.

［7］ Li Liyuan, Maglor K H. Integrating intensity and texture differences for robust change detection［J］.IEEE Trans Image Procession,2002,11(2):102-112.

［8］ Zhong J, Sclaroff S. Segmenting foreground objects from a dynamic textured background via a robust kalman filter ［C］ ∥Proceedings of International Conference on Computer Vision. USA:IEEE,2003：44-50.

［9］ Jabri S, Duric Z, Wechsler H. Detection and location people in video images using adaptive fusion of color and edge information ［C］∥Proceedings of International Conference on Pattern Recognition. Barcelona, 2000, 627-630.

［10］ Ercan Ozyildiz, Nils Krahnstover, Rajeev Sharma. Adaptive texture and color segmentation for tracking moving objects ［J］. Pattern Recognition, 2002, 35(10): 2013-2029.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed