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) ›› 2019, Vol. 49 ›› Issue (2): 116-121.doi: 10.6040/j.issn.1672-3961.0.2018.243

• 机器学习与数据挖掘 • 上一篇    下一篇

基于视频统计特征的差错敏感度模型

李童1,2(),马然1,2,*(),郑鸿鹤1,2,安平1,2,胡翔宇1,2   

  1. 1. 上海先进通信与数据科学研究院, 上海 200444
    2. 上海大学通信与信息工程学院, 上海 200444
  • 收稿日期:2018-05-25 出版日期:2019-04-20 发布日期:2019-04-19
  • 通讯作者: 马然 E-mail:lynne_li@shu.edu.cn;maran@shu.edu.cn
  • 作者简介:李童(1995—)，女，江西赣州人，硕士研究生，主要研究方向为图像与视频信号处理. E-mail: lynne_li@shu.edu.cn
  • 基金资助:
    国家自然科学基金项目(61301112)

An error sensitivity model based on video statistical features

Tong LI1,2(),Ran MA1,2,*(),Honghe ZHENG1,2,Ping AN1,2,Xiangyu HU1,2   

  1. 1. Shanghai Institute for Advanced Communication and Data Science, Shanghai 200444, Shanghai, China
    2. School of Communication and Information Engineering, Shanghai University, Shanghai 200444, Shanghai, China
  • Received:2018-05-25 Online:2019-04-20 Published:2019-04-19
  • Contact: Ran MA E-mail:lynne_li@shu.edu.cn;maran@shu.edu.cn
  • Supported by:
    国家自然科学基金项目(61301112)

RichHTML

5

PDF (PC)

374

摘要：

针对传统的丢包对视频质量影响研究的局限性,提出一种差错敏感度模型。对每个受损块提取周围块的丢失情况、纹理复杂度、运动矢量和梯度等可用的统计特征;对丢包视频进行差错隐藏,计算出差错敏感度;利用机器学习技术,建立统计特征和差错敏感度的关系模型。试验结果表明,相比于现有评价方法,该模型可以比较准确地预测视频帧局部差异性对不同丢包情况的敏感程度,尤其对于运动缓慢的视频序列,预测精度明显优于其他方法。

关键词: 丢包, 视频质量, 统计特征, 差错敏感度, 机器学习

Abstract:

The traditional packet losses affected the video quality, an error sensitivity model was proposed. For every damaged block, the available statistical features around the block were extracted, which included the losing status of neighboring blocks, texture complexity, motion vector and gradient. After concealing the damaged videos with error concealment methods, error sensitivities were computed. The relationship model between statistical features and error sensitivities was finally established by machine learning technology. Experimental results demonstrated that the proposed model could accurately predict the sensitivities of video frames′ local differences to different packet loss cases, compared with the state-of-art assessment methods, especially for the slow-motion video sequences, the prediction accuracy could be obviously superior to other methods.

Key words: packet loss, video quality, statistical features, error sensitivity, machine learning

中图分类号: 

  • TP391

图1

差错敏感度示意图"

图2

差错敏感度算法流程框图"

图3

受损块与邻块关系图"

图4

运动矢量关系图"

表1

序列信息"

序列 分辨率 帧率/(帧·s-1)
Shark 1920×1088 25
PoznanStreet 1920×1088 25
GT_Fly 1920×1088 25
Newspaper 1024×768 30
BookArrival 1024×768 16.67
BQmall 832×480 60

表2

不同模型在6个序列上的性能比较"

算法 Shark PoznanStreet GT_Fly Newspaper BookArrival BQmall
PLCC SRCC PLCC SRCC PLCC SRCC PLCC SRCC PLCC SRCC PLCC SRCC
DIVINE 0.897 7 0.880 5 0.607 6 0.428 0 0.856 3 0.831 7 0.509 0 0.503 9 0.614 3 0.505 3 0.734 9 0.743 9
BRISQUE 0.895 0 0.881 7 0.561 8 0.427 7 0.849 3 0.806 9 0.569 3 0.514 3 0.606 3 0.508 0 0.709 6 0.726 4
SSEQ 0.891 1 0.869 3 0.409 3 0.344 2 0.815 9 0.781 0 0.445 0 0.411 5 0.517 1 0.470 8 0.636 2 0.634 7
NOREQI 0.899 0 0.877 0 0.578 7 0.434 9 0.840 8 0.820 3 0.584 8 0.490 7 0.598 9 0.565 6 0.687 6 0.704 8
Proposed 0.898 9 0.879 6 0.772 8 0.574 5 0.820 2 0.791 3 0.763 1 0.660 2 0.634 2 0.585 7 0.736 5 0.769 2

表3

跨序列测试结果"

算法 Shark BookArrival BQmall
PLCC SRCC PLCC SRCC PLCC SRCC
DIVINE 0.761 1 0.785 8 0.114 7 0.211 5 0.417 8 0.405 1
BRISQUE 0.773 5 0.717 3 0.231 8 0.233 0 0.089 8 -0.043 1
SSEQ 0.397 2 0.404 9 0.166 4 0.132 5 0.217 7 0.256 0
NOREQI 0.622 1 0.476 7 0.143 6 0.203 1 0.144 4 0.161 0
Proposed 0.814 4 0.799 1 0.554 3 0.484 2 0.560 0 0.574 7

图5

不同特征组合预测结果"

1 USMAN M, HE X, XU M, et al. Survey of error concealment techniques: research directions and open issues[C]//Picture Coding Symposium. Cairns, Australia: IEEE, 2015: 233-238.
2 WAN S, YANG F, XIE Z. Evaluation of video quality degradation due to packet loss[C]//International Symposium on Intelligent Signal Processing and Communication Systems. Chengdu, China: IEEE, 2010: 1-4.
3 CHEN N, JIANG X, WANG C, et al. Study on relationship between network video packet loss and video quality[C]//International Congress on Image and Signal Processing. Shanghai, China: IEEE, 2011: 282-286.
4 SAPUTRA Y M, HENDRAWAN. The effect of packet loss and delay jitter on the video streaming performance using H.264/MPEG-4 Scalable Video Coding[C]//International Conference on Telecommunication Systems Services and Applications. Denpasar-Bali, Indonesia: IEEE, 2016.
5 BONDZULIC B P , PAVLOVIC B Z , PETROVIC V S , et al. Performance of peak signal-to-noise ratio quality assessment in video streaming with packet losses[J]. Electronics Letters, 2016, 52 (6): 454- 456.
doi: 10.1049/el.2015.3784
6 UHRINA M, VACULIK M. The impact of bitrate and packet loss on the video quality of H.264/AVC compression standard[C]//International Conference on Telecommunications and Signal Processing. Prague, Czech Republic: IEEE, 2015: 1-6.
7 CHEN N, JIANG X, WANG C. Impact of packet loss distribution on the perceived IPTV video quality[C]//International Congress on Image and Signal Processing. Chongqing, China: IEEE, 2013: 38-42.
8 PAULIKS R, SLAIDINS I, TRETJAKS K, et al. Assessment of IP packet loss influence on perceptual quality of streaming video[C]//Asia Pacific Conference on Multimedia and Broadcasting. Bali, Indonesia: IEEE, 2015: 1-6.
9 刘河潮, 常义林, 元辉, 等. 一种网络丢包的无参考视频质量评估方法[J]. 西安电子科技大学学报, 2012, 39 (2): 29- 34.
doi: 10.3969/j.issn.1001-2400.2012.02.006
LIU Hechao , CHANG Yilin , YUAN hui , et al. No-reference video quality assessment over the IP network based on packet loss[J]. Journal of Xidian University, 2012, 39 (2): 29- 34.
doi: 10.3969/j.issn.1001-2400.2012.02.006
10 刘河潮, 杨付正, 常义林, 等. 考虑丢包特性的无参考网络视频质量评估模型[J]. 西安交通大学学报, 2012, 46 (2): 130- 134.
LIU Hechao , YANG Fuzheng , CHANG Yilin , et al. A no-reference assessment model for quality of networked video based on features of packets loss[J]. Journal of Xi'an Jiaotong University, 2012, 46 (2): 130- 134.
11 TANG S , ALFACE P R . Impact of random and burst packet losses on H.264 scalable video coding[J]. IEEE Transactions on Multimedia, 2014, 16 (8): 2256- 2269.
doi: 10.1109/TMM.2014.2348947
12 KORHONEN J . Study of the subjective visibility of packet loss artifacts in decoded video sequences[J]. IEEE Transactions on Broadcasting, 2018, 64 (2): 354- 366.
doi: 10.1109/TBC.2018.2832465
13 GAO P , PENG Q , WEI X . Analysis of pacet-loss-induced distortion in view synthesis prediction-based 3-D video coding[J]. IEEE Transactions on Image Processing, 2017, 26 (6): 2781- 2796.
doi: 10.1109/TIP.2017.2690058
14 HEWAGE C T E R , MARTINI M G , APPUHAMI H D . A study on the impact of compression and packet losses on rendered 3D views[J]. Three-Dimensional Image Processing(3DIP) and Applications Ⅱ, 2012, 8290, 82901D-1- 82901D-9.
doi: 10.1117/12.909164
15 HARALICK R M , SHANMUGAM K , DINSTEIN I . Textural features for image classification[J]. IEEE Transactions on Systems Man & Cybernetics, 1973, smc-3 (6): 610- 621.
16 CHANG C C , LIN C J . Libsvm:a library for support vector machines[J]. ACM Transactions on Intelligent Systems and Technology, 2011, 2 (3): 1- 27.
17 MOORTHY A K , BOVIK A C . Blind image quality assessment:from natural scene statistics to perceptual quality[J]. IEEE Transactions on Image Processing, 2011, 20 (12): 3350- 64.
doi: 10.1109/TIP.2011.2147325
18 MITTAL A , MOORTHY A K , BOVIK A C . No-reference image quality assessment in the spatial domain[J]. IEEE Transactions on Image Processing, 2012, 21 (12): 4695- 4708.
doi: 10.1109/TIP.2012.2214050
19 LIU L , LIU B , HUANG H , et al. No-reference image quality assessment based on spatial and spectral entropies[J]. Signal Processing Image Communication, 2014, 9 (8): 856- 863.
20 OSZUST M . No-reference image quality assessment using image statistics and robust feature descriptors[J]. IEEE Signal Processing Letters, 2017, 24 (11): 1656- 1660.
doi: 10.1109/LSP.2017.2754539
[1] 邹启杰,李昊宇,张汝波,裴腾达,刘艳. 自主驾驶的人机交互控制[J]. 山东大学学报 (工学版), 2019, 49(2): 23-33.
[2] 张冕,黄颖,梅海艺,郭毓. 基于Kinect的配电作业机器人智能人机交互方法[J]. 山东大学学报 (工学版), 2018, 48(5): 103-108.
[3] 赵英弘,何潇,周东华. 一类含有传感器故障的网络化系统容错估计[J]. 山东大学学报(工学版), 2017, 47(5): 71-78.
[4] 刘洋,刘博,王峰. 基于Parameter Server框架的大数据挖掘优化算法[J]. 山东大学学报(工学版), 2017, 47(4): 1-6.
[5] 魏波,张文生,李元香,夏学文,吕敬钦. 一种选择特征的稀疏在线学习算法[J]. 山东大学学报(工学版), 2017, 47(1): 22-27.
[6] 周旺,张晨麟,吴建鑫. 一种基于Hartigan-Wong和Lloyd的定性平衡聚类算法[J]. 山东大学学报(工学版), 2016, 46(5): 37-44.
[7] 孟令恒,丁世飞. 基于单静态图像的深度感知模型[J]. 山东大学学报(工学版), 2016, 46(3): 37-43.
[8] 刘杰, 杨鹏, 吕文生, 刘阿古达木, 刘俊秀. 基于气象因素的PM2.5质量浓度预测模型[J]. 山东大学学报(工学版), 2015, 45(6): 76-83.
[9] 郑毅, 朱成璋. 基于深度信念网络的PM2.5预测[J]. 山东大学学报(工学版), 2014, 44(6): 19-25.
[10] 谢琳1,殷熙尧2,李凡长3,吴佳3. 一种逆归结学习表示[J]. 山东大学学报(工学版), 2013, 43(4): 46-50.
[11] 孙甲冰1,2,张承进1*. 有丢包的随机不确定参数系统的最优融合滤波[J]. 山东大学学报(工学版), 2011, 41(6): 59-65.
[12] 何雪英1,2, 秦伟1, 尹义龙1*, 赵联征1,乔昊3. 基于机器学习的视频指纹识别[J]. 山东大学学报(工学版), 2011, 41(4): 29-33.
[13] 师夏阳,王宇飞,胡永健. 基于轮廓波变换的隐写分析算法[J]. 山东大学学报(工学版), 2011, 41(2): 75-79.
[14] 梁春林1,彭凌西2*. 基于免疫网络的无监督式分类算法[J]. 山东大学学报(工学版), 2010, 40(5): 82-86.
[15] 郭茂祖 邹权 李文滨 韩英鹏. 生物信息学中的学习问题[J]. 山东大学学报(工学版), 2009, 39(3): 1-6.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
[1] 李 侃 . 嵌入式相贯线焊接控制系统开发与实现[J]. 山东大学学报(工学版), 2008, 38(4): 37 -41 .
[2] 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 .
[3] 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 .
[4] 李可,刘常春,李同磊 . 一种改进的最大互信息医学图像配准算法[J]. 山东大学学报(工学版), 2006, 36(2): 107 -110 .
[5] 季涛,高旭,孙同景,薛永端,徐丙垠 . 铁路10 kV自闭/贯通线路故障行波特征分析[J]. 山东大学学报(工学版), 2006, 36(2): 111 -116 .
[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] 刘文亮,朱维红,陈涤,张泓泉. 基于雷达图像的运动目标形态检测及跟踪技术[J]. 山东大学学报(工学版), 2010, 40(3): 31 -36 .
[8] 孙炜伟,王玉振. 考虑饱和的发电机单机无穷大系统有限增益镇定[J]. 山东大学学报(工学版), 2009, 39(1): 69 -76 .
[9] 孙殿柱,朱昌志,李延瑞 . 散乱点云边界特征快速提取算法[J]. 山东大学学报(工学版), 2009, 39(1): 84 -86 .
[10] 杨发展1 ,艾兴1 ,赵军1 ,侯建锋2 . ZrO2含量对WC基复合材料的力学性能和微观结构的影响[J]. 山东大学学报(工学版), 2009, 39(1): 92 -95 .
版权所有 © 2018 《山东大学学报(工学版)》编辑部　
地址: 济南市山大南路27号（250100）　 电话: 0531-88366735　 E-mail: xbgxb@sdu.edu.cn
本系统由北京玛格泰克科技发展有限公司设计开发