Please wait a minute...
Journal of ZheJiang University (Engineering Science)  2020, Vol. 54 Issue (7): 1249-1255    DOI: 10.3785/j.issn.1008-973X.2020.07.001
Dynamic image background modeling method for detecting abandoned objects in highway
Ying-jie XIA(),Cong-yu OUYANG
College of Computer Science and Technology, Zhejiang University, Hangzhou 310027, China
Download: HTML     PDF(1099KB) HTML
Export: BibTeX | EndNote (RIS)      


There have been some research in different image background modeling methods to detect abandoned objects in highway scenes. However, traditional fixed background modeling methods easily generate foreground noises because of the environmental changes, and dynamic background modeling methods quickly integrate the motionless foreground abandoned objects into the background model. A dynamic background modeling method was proposed based on background separation Gaussian mixture model (BS-GMM) for detecting abandoned objects in highway to solve this problem. Background division method and model matching method were improved in traditional Gaussian mixture model. The weight attenuation of the Gaussian distribution models per pixel was utilized to dynamically model and update image background model. The background update frequency of the traditional Gaussian mixture model method was retained, and the stationary target was continuously detected by the method of background separation. The method can reduce the impact of environmental noises easily generated in the open environment of highway, and effectively detect the long-time motionless abandoned objects. The method can achieve the effect of real-time detection in terms of computing performance. The experimental results show that our BS-GMM method produces less foreground noises than other methods, and detects abandoned objects which are motionless for more than 20 seconds. BS-GMM method can be effectively applied to detect abandoned objects in highway.

Key wordshighway      abandoned object detection      Gaussian mixture model      background modeling      weight attenuation     
Received: 25 February 2020      Published: 05 July 2020
CLC:  TP 391  
Cite this article:

Ying-jie XIA,Cong-yu OUYANG. Dynamic image background modeling method for detecting abandoned objects in highway. Journal of ZheJiang University (Engineering Science), 2020, 54(7): 1249-1255.

URL:     OR


针对高速公路抛洒物检测中传统的固定背景建模方法容易因开放环境变化产生大量的前景噪声,动态背景建模方法容易因抛洒物的静止特性导致前景对象快速融入背景,提出基于背景分离高斯混合模型(BS-GMM)的动态背景建模方法. 该方法对传统高斯混合模型的背景划分和模型匹配方法进行改进,设计基于像素点的高斯分布背景模型权值的衰减状况进行背景建模和背景更新的方法,既能减少开放环境大量环境噪声的影响,也能对抛洒物快速进入静止状态后的准确检测,在计算性能上能够达到实时检测的效果. 实验结果证明,BS-GMM方法在抛洒物检测过程中产生的噪声数量比其他方法少,且对静止超过20 s的物体能够作为前景目标提取,因此能够有效地应用于高速公路抛洒物的准确识别.

关键词: 高速公路,  抛洒物检测,  高斯混合模型,  背景建模,  权值衰减 
Fig.1 BS-GMM method flow chart
Fig.2 Traditional background subtraction
Fig.3 Improved background subtraction
视频类型 场景数量 视频数量 视频时长/min
含抛洒物 30 40 2
无抛洒物 30 120 2
Tab.1 Experimental data set of video
Fig.4 Correctness of abandoned objects detection
Fig.5 Effect of static abandoned object sustainable extraction
Fig.6 Number of noise in foreground of single video
Fig.7 Effect of foreground extraction in varying environmental conditions
Fig.8 Average number of noise in foreground of experimental data set
Fig.9 Effect of abandoned objects detection
[1]   叶芳芳, 许力 实时的静止目标与鬼影检测及判别方法[J]. 浙江大学学报: 工学版, 2015, 49 (1): 181- 185
YE Fang-fang, XU Li Real-time detection and discrimination of static objects and ghosts[J]. Journal of Zhejiang University: Engineering Science, 2015, 49 (1): 181- 185
[2]   BEGEL A, PHANG K Y, ZIMMERMANN T. Codebook: discovering and exploiting relationships in software repositories [C] // IEEE International Conference on Software Engineering. Cape Town: IEEE, 2010: 125-134.
[3]   STAUFFER C. Adaptive background mixture model for real-time tracking [C] // IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Fort Collins: IEEE, 1998: 246-252.
[4]   陈祖爵, 陈潇君, 何鸿 基于改进的混合高斯模型的运动目标检测[J]. 中国图象图形学报, 2007, 12 (9): 1585- 1589
CHEN Zu-jue, CHEN Xiao-jun, HE Hong Moving object detection based on improved mixture Gaussian models[J]. Journal of Image and Graphics, 2007, 12 (9): 1585- 1589
doi: 10.3969/j.issn.1006-8961.2007.09.014
[5]   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]. 中国图象图形学报, 2008, 13 (4): 729- 734
LIU Xin, LIU Hui, QIANG Zhen-ping, et al Adaptive background modeling based on mixture Gaussian model and frame subtraction[J]. Journal of Image and Graphics, 2008, 13 (4): 729- 734
doi: 10.11834/jig.20080422
[6]   SILVIA F, GIANLUCA G, MASSIMO N, et al. A new method for real time abandoned object detection and owner tracking [C] // 2006 IEEE International Conference on Image Processing. Atlanta: IEEE, 2006: 3329-3332.
[7]   汪贵平, 马力旺, 郭璐, 等 高速公路抛洒物事件图像检测算法[J]. 长安大学学报: 自然科学版, 2017, 37 (5): 81- 88
WANG Gui-ping, MA Li-wang, GUO Lu, et al Image detection algorithm for incident of discarded things in highway[J]. Journal of Chang'an University: Natural Science Edition, 2017, 37 (5): 81- 88
[8]   FU H, MEI X, MAL H, et al. Abandoned object detection in highway scene [C] // 6th International Conference on Pervasive Computing and Applications. Port Elizabeth: IEEE, 2011: 117-121.
[9]   WEN J, GONG H, ZHANG X, et al. Generative model for abandoned object detection [C] // Proceedings of the International Conference on Image Processing. Cairo: IEEE, 2009: 853-856.
[10]   ELENA S, CARLO S R Real-time video-shot detection for scene surveillance applications[J]. IEEE Transactions on Image Processing A Publication of the IEEE Signal Processing Society, 2000, 9 (1): 69- 79
doi: 10.1109/83.817599
[11]   阮雅端, 高妍, 张宇杭, 等. 基于深度卷积网络的路面遗留物检测方法技术: 201710059678.0 [P]. 2017-06-13.
RUAN Ya-duan, GAO Yan, ZHANG Yu-hang, et al. Detection method technology of pavement residues based on deep convolution network: 201710059678.0 [P]. 2017-06-13.
[12]   金瑶, 张锐, 尹东 城市道路视频中小像素目标检测[J]. 光电工程, 2019, 46 (9): 76- 83
JIN Yao, ZHANG Rui, YIN Dong Object detection for small pixel in urban roads videos[J]. Opto-Electronic Engineering, 2019, 46 (9): 76- 83
[13]   ZIVKOVIC Z. Improved adaptive Gaussian mixture model for background subtraction [C] // Proceedings of the 17th International Conference on Pattern Recognition. Cambridge: IEEE, 2004: 28-31.
[14]   YINGDONG M A, LIU Y, LIU S, et al Multiple object detection and tracking in complex background[J]. International Journal of Pattern Recognition and Artificial Intelligence, 2016, 31 (2): 809- 830
[15]   赵占杰, 林小竹, 张金燕 一种基于码本背景模型的运动目标检测方法[J]. 北京石油化工学院学报, 2010, (1): 45- 49
ZHAO Zhan-jie, LIN Xiao-zhu, ZHANG Jin-yan A moving object detection method based on codebook background model[J]. Journal of Beijing Institute of Petro-chemical Technology, 2010, (1): 45- 49
[1] Huang-yao ZENG,Dan-dan LI,Yan MA,Qun CONG. Risky accounts evaluation method of campus network[J]. Journal of ZheJiang University (Engineering Science), 2020, 54(9): 1761-1767.
[2] Chao SUN,Meng-hui LI,Fei HAN. Traffic evolution model with multi-source data of intelligent highway[J]. Journal of ZheJiang University (Engineering Science), 2020, 54(3): 546-556.
[3] Fen-jie LONG,Long-fei ZHENG,Lang SHI,Zi-yan CHEN. Regional integration of Qianzhong cities using highway big data[J]. Journal of ZheJiang University (Engineering Science), 2019, 53(2): 307-314.
[4] XU Cheng, QU Zhao-wei, WANG Dian-hai, JIN Sheng. Speed distribution model for heterogeneous bicycle traffic flow[J]. Journal of ZheJiang University (Engineering Science), 2017, 51(7): 1331-1338.
[5] CHEN De, HAN Sen, SU Qian, HAN Xiao. Evaluation indicator of surface texture of asphalt pavement based on skid-resistance and noise reduction performance[J]. Journal of ZheJiang University (Engineering Science), 2017, 51(5): 896-903.
[6] ZHAO Xue-wu, JI Jun-zhong, YAO Yao. Insula functional parcellation by searching Gaussian mixture model (GMM) using immune clonal selection (ICS) algorithm[J]. Journal of ZheJiang University (Engineering Science), 2017, 51(12): 2320-2331.
[7] CHAN Zhen-Yu, YANG Ying-Chun. Universal background model reduction based efficient speaker recognition[J]. Journal of ZheJiang University (Engineering Science), 2009, 43(6): 978-982.