您的位置:山东大学 -> 科技期刊社 -> 《山东大学学报(工学版)》

山东大学学报 (工学版) ›› 2020, Vol. 50 ›› Issue (3): 38-44.doi: 10.6040/j.issn.1672-3961.0.2019.413

• 机器学习与数据挖掘 • 上一篇    下一篇

基于背景复杂度自适应距离阈值的修正SuBSENSE算法

成科扬1,2(),孙爽1,詹永照1   

  1. 1. 江苏大学计算机科学与通信工程学院,江苏 镇江 212013
    2. 社会安全风险感知与防控大数据应用国家工程实验室,北京 100846
  • 收稿日期:2019-07-22 出版日期:2020-06-20 发布日期:2020-06-16
  • 作者简介:成科扬(1982—),男,江苏南通人,副教授,博士,主要研究方向为模式识别,计算机视觉. E-mail: kycheng@ujs.edu.cn
  • 基金资助:
    国家自然科学基金资助项目(61972183);国家自然科学基金资助项目(61602215);社会安全风险感知与防控大数据应用国家工程实验室主任基金项目

Modified SuBSENSE algorithm via adaptive distance threshold based on background complexity

Keyang CHENG1,2(),Shuang SUN1,Yongzhao ZHAN1   

  1. 1. School of Computer Science and Telecommunications Engineering, Jiangsu University, Zhenjiang 212013, Jiangsu, China
    2. National Engineering Laboratory for Public Safety Risk Perception and Control by Big Data, Beijing 100846, China
  • Received:2019-07-22 Online:2020-06-20 Published:2020-06-16
  • Supported by:
    国家自然科学基金资助项目(61972183);国家自然科学基金资助项目(61602215);社会安全风险感知与防控大数据应用国家工程实验室主任基金项目

摘要:

针对自适应敏感度分割(self-balanced sensitivity segmenter, SuBSENSE)算法在真实复杂场景下距离阈值更新适应性差,导致检测效果不佳的问题,提出一种基于背景复杂度自适应距离阈值修正的SuBSENSE算法。结合时间一致性和空间一致性定义了一种背景复杂度的度量方式,以此为标准,通过距离阈值修正策略获取准确的距离阈值,以便获得更好的检测效果。本算法与像素自适应分割(based adaptive segmenter,PBAS)算法和传统SuBSENSE算法进行了对比。试验表明,在动态场景下,本算法获取的前景更加精确,精度比PBAS算法和传统SuBSENSE算法提高了6.70%和0.80%,召回率比PBAS算法和传统SuBSENSE算法分别提高了9.37%和1.24%。本算法优于对比算法,在动态场景下具有更高的鲁棒性和检测精度。

关键词: SuBSENSE算法, 前景检测, 距离阈值修正, 背景复杂度

Abstract:

In order to solve the problem of poor adaptability of SuBSENSE algorithm in updating distance threshold in real complex scenes, which resulted in poor detection effect, SuBSENSE algorithm is proposed based on adaptive distance threshold correction of background complexity. A measure of background complexity is defined based on temporal consistency and spatial consistency, and the distance threshold correction strategy to get the accurate distance threshold as a criterion to achieve better detection results. This algorithm was compared with PBAS and traditional SuBSENSE algorithm. Experiments showed that the prospects of the proposed algorithm were more accurate in dynamic scenarios. The precision of the proposed algorithm was 6.70% and 0.80% higher than that of the PBAS algorithm and the traditional SuBSENSE algorithm, and the recall was 9.37% and 1.24% higher than that of the PBAS algorithm and the traditional SuBSENSE algorithm, respectively. After a comprehensive study of the three indicators, it was found that the proposed algorithm was superior to the contrast algorithms, and had higher robustness and detection accuracy in dynamic scenarios.

Key words: SuBSENSE algorithm, foreground detection, modified distance threshold, background complexity

中图分类号: 

  • TP391

图1

PBAS算法、SuBSENSE算法和本研究算法的对比试验"

表1

召回率指标对比结果"

dataset PBAS SuBSENSE Ours
overpass 67.04 78.52 79.07
canoe 56.25 65.90 67.62
fountain01 86.38 87.71 89.13
fountain02 90.77 92.32 93.48
fall 94.74 85.67 85.67
boats 22.13 55.96 58.53
overall 69.55 77.68 78.92

表2

精度指标对比结果"

dataset PBAS SuBSENSE Ours
overpass 96.90 94.37 94.98
canoe 99.86 99.33 99.45
fountain01 27.51 65.99 67.64
fountain02 96.51 96.58 96.89
fall 80.67 87.58 88.14
boats 98.08 91.06 92.63
overall 83.26 89.15 89.96

表3

F1指标对比结果"

dataset PBAS SuBSENSE Ours
overpass 79.25 85.72 86.30
canoe 71.96 79.23 80.50
fountain01 41.73 75.31 76.91
fountain02 93.55 94.41 95.16
fall 87.14 86.61 86.89
boats 36.11 69.32 71.73
overall 68.29 81.77 82.92

图2

3种算法不同评价指标对比"

表4

各算法综合对比"

Methods Recall/
%
Precise/
%
F-Measure/
%
处理帧率/
(帧·s-1)
PBAS 69.55 83.26 68.29 37
SuBSENSE 77.68 89.15 81.77 30
Ours 78.92 89.96 82.92 27
1 JAIN R , NAGEL H H . On the analysis of accumulative difference pictures from image sequences of real world scenes[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1978, PAMI-1 (2): 206- 214.
2 BARRON J L , FLEET D J , BEAUCHEMIN S S , et al. Performance of optical flow techniques[J]. International Journal of Computer Vision, 1994, 12 (1): 43- 77.
3 BEAUGENDRE A , GOTO S . Block-propagative background subtraction system for UHDTV videos[J]. IPSJ Transactions on Computer Vision and Applications, 2015, 7, 31- 34.
doi: 10.2197/ipsjtcva.7.31
4 MAITY S, CHAKRABARTI A, BHATTACHARJEE D. Block-based quantized histogram (BBQH) for efficient background modeling and foreground extraction in video[C]//2017 Inter-national Conference on Data Management, Analytics and Innovation. Pune, India: IEEE, 2017: 224-229.
5 BARNICH O , VAN D M . ViBe: a universal background subtraction algorithm for video sequences[J]. IEEE Transactions on Image Processing, 2011, 20 (6): 1709- 1724.
doi: 10.1109/TIP.2010.2101613
6 惠开发, 成科扬, 詹永照. 基于改进ViBe算法的视频浓缩[J]. 山东大学学报(工学版), 2017, 47 (3): 43- 48.
HUI Kaifa , CHENG Keyang , ZHAN Yongzhao . The video sy-nopsis based on the enhanced ViBe algorithm[J]. Journal of Shandong University (Engineering Science), 2017, 47 (3): 43- 48.
7 HOFMANN M, TIEFENBACHER P, RIGOLL G. Backg-round segmentation with feedback: the pixel-based adaptive segmenter[C]//IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops. New York, USA: IEEE, 2012: 38-43.
8 ST-CHARLES P L, BILODEAU G A. Improving background subtraction using local binary similarity patterns[C]// 2014 IEEE Winter Conference on Applications of Computer Vision. Steamboat Springs, USA: IEEE, 2014: 509-515.
9 陈树, 丁保阔. 动态背景下自适应LOBSTER算法的前景检测[J]. 中国图象图形学报, 2017, 22 (2): 161- 169.
CHEN Shu , DING Baokuo . Foreground detection of the adaptive LOBSTER algorithm in a dynamic background[J]. Journal of Image and Graphics, 2017, 22 (2): 161- 169.
10 杨丹, 戴芳. 运动目标检测的ViBe算法改进[J]. 中国图象图形学报, 2018, 23 (12): 1813- 1828.
doi: 10.11834/jig.180304
YANG Dan , DAI Fang . Improvement of ViBe algorithm for moving target detection[J]. Journal of Image and Graphics, 2018, 23 (12): 1813- 1828.
doi: 10.11834/jig.180304
11 GUO Lili, XU Dan, QIANG Zhenping. Background subtraction using local svd binary pattern[C]//IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops. Las Vegas, USA: IEEE, 2016.
12 ST-CHARLES P L , BILODEAU G A , BERGEVIN R . SuBSENSE: a universal change detection method with local adaptive sensitivity[J]. IEEE Transactions on Image Processing, 2015, 24 (1): 359- 373.
doi: 10.1109/TIP.2014.2378053
13 李静, 刘清, 颜为朗. 基于SuBSENSE的内河船舶检测波纹干扰抑制算法[J]. 交通信息与安全, 2017, 35 (2): 30- 34.
LI Jing , LIU Qing , YAN Weilang . An algorithm for ripple suppression of inland ship detection based on SuBSENSE[J]. Journal of Transport Information and Safety, 2017, 35 (2): 30- 34.
14 ST-CHARLES P L, BILODEAU G A, BERGEVIN R. A self-adjusting approach to change detection based on background word consensus[C]//2015 IEEE Winter Conference on Applications of Computer Vision. Waikoloa HI, USA: IEEE, 2015: 990-997.
15 ST-CHARLES P L , BILODEAU G A , BERGEVIN R . Universal background subtraction using word consensus models[J]. IEEE Transactions on Image Processing, 2016, 25 (10): 4768- 4781.
doi: 10.1109/TIP.2016.2598691
16 LUO Huaiye, LI Bo, ZHOU Zhiheng. Improved background subtraction based on word consensus models[C]//2017 International Symposium on Intelligent Signal Processing and Communication Systems. Xiamen, China: IEEE, 2017: 689-694.
17 汪荣琪, 郑林, 王标. 基于改进的PBAS算法的前景目标检测[J]. 计算机科学, 2017, 44 (5): 294- 298.
WANG Rongqi , ZHENG Lin , WANG Biao . Foreground object detection based on improved PBAS[J]. Computer Science, 2017, 44 (5): 294- 298.
18 钟忺, 汪梦, 张倩, 等. 一种基于纹理和颜色置信融合的运动目标检测方法[J]. 计算机应用研究, 2017, 34 (7): 2196- 2201.
doi: 10.3969/j.issn.1001-3695.2017.07.060
ZHONG Wei , WANG Meng , ZHANG Qian , et al. Moving object detection by fusing texture and color features with confidence[J]. Application Research of Computers, 2017, 34 (7): 2196- 2201.
doi: 10.3969/j.issn.1001-3695.2017.07.060
19 WANG Yi, JODOIN P M, PORIKL F, et al. CDnet2014: an expanded change detection benchmark dataset[C]//IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops. Columbus, USA: IEEE, 2014.
20 郭丽丽.面向复杂场景的运动目标检测技术研究[D].昆明:云南大学, 2017.
GUO Lili. Research on techniques of moving object detection under complex environment[D]. Kunming: Yunnan University, 2017.
[1] 田枫,李欣,刘芳,李闯,孙小强,杜睿山. 基于多模态子空间学习的语义标签生成方法[J]. 山东大学学报 (工学版), 2020, 50(3): 31-37, 44.
[2] 马金平. 基于UART串口的多机通讯[J]. 山东大学学报 (工学版), 2020, 50(3): 24-30.
[3] 袁高腾,刘毅慧,黄伟,胡兵. 基于Gabor特征的乳腺肿瘤MR图像分类识别模型[J]. 山东大学学报 (工学版), 2020, 50(3): 15-23.
[4] 段江丽,胡新. 自然语言问答中的语义关系识别[J]. 山东大学学报 (工学版), 2020, 50(3): 1-7.
[5] 严云洋,杜晨锡,刘以安,高尚兵. 基于轻型卷积神经网络的火焰检测方法[J]. 山东大学学报 (工学版), 2020, 50(2): 100-107.
[6] 张胜男,王雷,常春红,郝本利. 基于三维剪切波变换和BM4D的图像去噪方法[J]. 山东大学学报 (工学版), 2020, 50(2): 83-90.
[7] 胡龙茂,胡学钢. 基于多维相似度和情感词扩充的相同产品特征识别[J]. 山东大学学报 (工学版), 2020, 50(2): 50-59.
[8] 陈艳平,冯丽,秦永彬,黄瑞章. 一种基于深度神经网络的句法要素识别方法[J]. 山东大学学报 (工学版), 2020, 50(2): 44-49.
[9] 闫威,张达敏,张绘娟,辛梓芸,陈忠云. 基于混合决策的改进鸟群算法[J]. 山东大学学报 (工学版), 2020, 50(2): 34-43.
[10] 宋士奇,朴燕,蒋泽新. 基于改进YOLOv3的复杂场景车辆分类与跟踪[J]. 山东大学学报 (工学版), 2020, 50(2): 27-33.
[11] 陈宁宁,赵建伟,周正华. 基于校正神经网络的视频追踪算法[J]. 山东大学学报 (工学版), 2020, 50(2): 17-26.
[12] 赵越男,陈桂友,孙琛,卢宁,廖立伟. 基于空间隐患分布与运动意图解析的危险评估方法[J]. 山东大学学报 (工学版), 2020, 50(1): 28-34.
[13] 苏佳林,王元卓,靳小龙,程学旗. 自适应属性选择的实体对齐方法[J]. 山东大学学报 (工学版), 2020, 50(1): 14-20.
[14] 蔡国永,林强,任凯琪. 基于域对抗网络和BERT的跨领域文本情感分析[J]. 山东大学学报 (工学版), 2020, 50(1): 1-7,20.
[15] 姚元玺. 基于分场景重构的风电汇聚趋势性量化方法[J]. 山东大学学报 (工学版), 2019, 49(6): 86-92.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
[1] 施来顺,万忠义 . 新型甜菜碱型沥青乳化剂的合成与性能测试[J]. 山东大学学报(工学版), 2008, 38(4): 112 -115 .
[2] 韩雪. 平庄西露天煤矿滑坡灾害远程监测实例分析[J]. 山东大学学报(工学版), 2009, 39(4): 116 -120 .
[3] 余嘉元1 , 田金亭1 , 朱强忠2 . 计算智能在心理学中的应用[J]. 山东大学学报(工学版), 2009, 39(1): 1 -5 .
[4] 李梁,罗奇鸣,陈恩红. 对象级搜索中基于图的对象排序模型(英文)[J]. 山东大学学报(工学版), 2009, 39(1): 15 -21 .
[5] 陈瑞,李红伟,田靖. 磁极数对径向磁轴承承载力的影响[J]. 山东大学学报(工学版), 2018, 48(2): 81 -85 .
[6] 浦剑1 ,张军平1 ,黄华2 . 超分辨率算法研究综述[J]. 山东大学学报(工学版), 2009, 39(1): 27 -32 .
[7] 秦通,孙丰荣*,王丽梅,王庆浩,李新彩. 基于极大圆盘引导的形状插值实现三维表面重建[J]. 山东大学学报(工学版), 2010, 40(3): 1 -5 .
[8] Yue Khing Toh1 , XIAO Wendong2 , XIE Lihua1 . 基于无线传感器网络的分散目标跟踪:实际测试平台的开发应用(英文)[J]. 山东大学学报(工学版), 2009, 39(1): 50 -56 .
[9] 岳远征. 远离平衡态玻璃的弛豫[J]. 山东大学学报(工学版), 2009, 39(5): 1 -20 .
[10] 田芳1,张颖欣2,张礼3,侯秀萍3,裘南畹3. 新型金属氧化物薄膜气敏元件基材料的开发[J]. 山东大学学报(工学版), 2009, 39(2): 104 -107 .