基于结构特征约束两阶段重构的视频压缩感知

doi:10.3969/j.issn.1001-506X.2020.11.05

系统工程与电子技术 ›› 2020, Vol. 42 ›› Issue (11): 2441-2449.doi: 10.3969/j.issn.1001-506X.2020.11.05

基于结构特征约束两阶段重构的视频压缩感知

刘歌(), 芮国胜(), 田文飚()

海军航空大学信号与信息处理山东省重点实验室, 山东烟台 264001

收稿日期:2020-02-16 出版日期:2020-11-01 发布日期:2020-11-05
作者简介:刘歌(1991-),女,博士研究生,主要研究方向为压缩感知、蒸发波导反演。E-mail:yyliuge@sina.com|芮国胜(1968-),男,教授,博士研究生导师,博士,主要研究方向为压缩感知、现代滤波理论。E-mail:ruigs@vip.sina.com|田文飚(1987-),男,副教授,博士,主要研究方向为压缩感知、蒸发波导反演。E-mail:twbi5si@gmail.com
基金资助:
国家自然科学基金(41606117);国家自然科学基金(41476089);国家自然科学基金(61671016)

Video compressed sensing based on two-phase reconstruction of structural feature prior constraints

Ge LIU(), Guosheng RUI(), Wenbiao TIAN()

Signal and Information Processing Provincial Key Laboratory in Shandong, Naval Aviation University, Yantai 264001, China

Received:2020-02-16 Online:2020-11-01 Published:2020-11-05

摘要/Abstract

摘要：

针对现有视频压缩感知多假设预测-残差重构方法重构精度不高的问题,提出一种基于结构特征先验约束两阶段重构的多假设预测视频压缩感知方法。该方法从相似图像块非局部相似性和梯度稀疏性出发,将第一阶段多假设预测重构后的当前帧直接作为第二阶段重构的初始重构帧,利用低秩正则化和全变差正则化再次进行重构,其中低秩正则化矩阵是通过欧氏距离-感知哈希算法获取的图像相似块集合,同时包含帧内和帧间的相似图像块,充分利用帧内帧间的结构相似性,有效提高重构性能,为后续残差重构打下基础。仿真实验表明,所提两阶段重构算法较现有几种优秀重构算法更好地保留了视频帧的细节,并具有更高的重构精度。

关键词: 视频压缩感知, 多假设预测, 结构先验, 低秩性, 全变差

Abstract:

To solve the problem that the accuracy of existing multi-hypothesis prediction methods for compressed video sensing is not high, a multi-hypothesis prediction method based on structural feature priori constrained two-phase reconstruction is proposed. Starting from the non-local similarity and the gradient sparsity of the similar image blocks, the current frame reconstructed by the multi-hypothesis prediction is directly used as the initial frame of the second phase reconstruction. The low rank regularization and the total variation regularization are utilized to perform the second phase reconstruction, wherein the low rank regularization matrix includes the similar blocks of intra-frames and inter-frames, fully utilizing the structural similarity between intra-frames, and improving reconstruction performance effectively, laying the foundation for the subsequent reconstruction of residuals. Simulation experiments show that the proposed two-stage reconstruction algorithm preserves the details of video frames better and has a higher reconstruction accuracy than several excellent reconstruction algorithms.

Key words: video compressed sensing, multi-hypothesis prediction, structure prior, low rank, total variation

中图分类号:

TN911.73

刘歌, 芮国胜, 田文飚. 基于结构特征约束两阶段重构的视频压缩感知[J]. 系统工程与电子技术, 2020, 42(11): 2441-2449.

Ge LIU, Guosheng RUI, Wenbiao TIAN. Video compressed sensing based on two-phase reconstruction of structural feature prior constraints[J]. Systems Engineering and Electronics, 2020, 42(11): 2441-2449.

图/表 8

图1

图2

图3

图4

图5

表1

图6

图7

参考文献 19

1	DONOHO D L . Compressed sensing[J]. IEEE Trans.on Information Theory, 2006, 52 (4): 1289- 1306. doi: 10.1109/TIT.2006.871582
2	RANI M , DHOK S B , DESHMUKH R B . A systematic review of compressive sensing: concepts, implementations and applications[J]. IEEE Access, 2018, 6, 4875- 4894. doi: 10.1109/ACCESS.2018.2793851
3	BARANIUK R G , GOLDSTEIN T , SANKARANARAYANAN A C , et al. Compressive video sensing: algorithms, architectures, and applications[J]. IEEE Signal Processing Magazine, 2017, 34 (1): 52- 66. doi: 10.1109/MSP.2016.2602099
4	DRORI I . Compressed video sensing[J]. Bmva Symposium Ond Video, 2008, 11 (4): 1- 2.
5	COSSALTER M , VALENZISE G , TAGLIASACCHI M , et al. Joint compressive video coding and analysis[J]. IEEE Trans.on Multimedia, 2010, 12 (3): 168- 183. doi: 10.1109/TMM.2010.2041105
6	ZHAO C, MA S W, GAO W. Video compressive sensing via structured Laplacian modelling[C]//Proc.of the IEEE Visual Communications and Image Processing Conference, 2015: 402-405.
7	LIU Y , LI M , PADOS D A . Motion-aware decoding of compressed-sensed video[J]. IEEE Trans.on Circuits & Systems for Video Technology, 2013, 23 (3): 438- 444.
8	YANG J B , YUAN X , LIAO X J , et al. Video compressive sensing using Gaussian mixture models[J]. IEEE Trans.on Image Processing, 2014, 23 (11): 4863- 4878. doi: 10.1109/TIP.2014.2344294
9	CHEN C, TRAMEL E W, FOWLER J E. Compressed-sensing recovery of images and video using multi-hypothesis predictions[C]//Proc.of the Signals, Systems & Computers, 2011: 1193-1198.
10	CHEN J , CHEN Y Z , QIN D , et al. An elastic net-based hybrid hypothesis method for compressed video sensing[J]. Multimedia Tools and Applications, 2015, 74 (6): 2085- 2108. doi: 10.1007/s11042-013-1743-y
11	AZGHANI M , KARIMI M , MARVASTU F . Multi-hypothesis compressed video sensing technique[J]. IEEE Trans.on Circuits & Systems for Video Technology, 2014, 26 (4): 627- 635.
12	CHEN J , WANG N , XUE F , et al. Distributed compressed video sensing based on the optimization of hypothesis set update technique[J]. Multimedia Tools & Applications, 2017, 76 (14): 15735- 15754.
13	ZHAO C , MA S W , ZHANG J , et al. Video compressive sensing reconstruction via reweighted residual sparsity[J]. IEEE Trans.on Circuits and Systems for Video Technology, 2017, 27 (6): 1182- 1195. doi: 10.1109/TCSVT.2016.2527181
14	NIU X M , JIAO Y H . An overview of perceptual hashing[J]. Acta Electronica Sinica, 2008, 36 (7): 1405- 1411.
15	XU J , QIAO Y S , FU Z Z , et al. Image block compressive sensing reconstruction via group-based sparse representation and nonlocal total variation[J]. Circuits, systems, and signal processing, 2019, 38 (1): 304- 328.
16	ZHA Z Y , ZHANG X G , WANG Q , et al. Group-based sparse representation for image compressive sensing reconstruction with non-convex regularization[J]. Neurocomputing, 2018, 296, 55- 63. doi: 10.1016/j.neucom.2018.03.027
17	DONG W , SHI G , LI X , et al. Compressive sensing via nonlocal low-rank regularization[J]. IEEE Trans.on Image Processing, 2014, 23 (8): 3618- 3632. doi: 10.1109/TIP.2014.2329449
18	BOYD S , PARIKH N , CHU E , et al. Distributed optimization and statistical learning via the alternating direction method of multipliers[J]. Foundations and Trends in Machine Learning, 2011, 3 (1): 1- 122.
19	杨春玲, 欧伟枫. CVS中基于多参考帧的最优多假设预测算法[J]. 华南理工大学学报(自然科学版), 2016, 44 (1): 1- 8.
	YANG C L , OU W F . Multi-reference frames-based optimal multi-hypothesis prediction algorithm for compressed video sensing[J]. Journal of South China University of Technology (Natural Science ed.), 2016, 44 (1): 1- 8.

视频序列	重构算法	PSNR/dB
视频序列	重构算法	0.1	0.2	0.3	0.4	0.5	平均值
Foreman	MH-Tikhonov	31.80	33.58	37.35	38.39	39.10	36.64
	MH-wElasticNet	32.03	34.90	37.11	39.21	41.22	36.89
	HD-BDCVS	33.62	36.23	37.88	39.52	40.88	37.63
	MH-RSS	33.50	36.83	38.21	41.09	42.83	38.49
	MH-SFC	33.63	37.03	39.42	41.86	43.65	39.12
Hall	MH-Tikhonov	31.05	32.45	33.89	35.04	36.25	33.74
	MH-wElasticNet	32.18	33.84	35.08	36.12	37.28	34.90
	HD-BDCVS	33.02	34.12	35.53	36.52	38.46	35.53
	MH-RSS	33.43	34.79	35.55	36.76	38.07	35.72
	MH-SFC	34.23	35.32	36.96	37.11	38.25	36.37
Football	MH-Tikhonov	25.24	26.85	27.96	29.60	30.98	28.13
	MH-wElasticNet	25.94	27.31	28.95	30.92	31.80	28.98
	HD-BDCVS	26.09	27.86	29.26	31.06	32.43	29.34
	MH-RSS	27.21	28.02	29.13	31.12	32.68	29.63
	MH-SFC	27.89	28.85	29.91	31.84	33.42	30.38
Coastguard	MH-Tikhonov	27.44	29.54	31.21	32.16	33.78	30.83
	MH-wElasticNet	28.24	30.35	32.19	33.81	35.54	32.03
	HD-BDCVS	30.15	32.41	34.23	36.45	38.10	34.27
	MH-RSS	30.33	32.31	34.06	35.58	37.16	33.89
	MH-SFC	30.69	32.52	35.02	36.75	38.16	34.63
News	MH-Tikhonov	33.12	35.21	37.17	38.36	39.49	36.67
	MH-wElasticNet	34.24	35.35	38.19	39.81	40.54	37.63
	HD-BDCVS	35.15	36.41	39.23	40.45	41.10	38.47
	MH-RSS	35.28	36.31	39.06	40.58	41.16	38.48
	MH-SFC	36.69	37.12	39.02	41.15	41.57	39.11

基于结构特征约束两阶段重构的视频压缩感知

Video compressed sensing based on two-phase reconstruction of structural feature prior constraints

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 8

参考文献 19

相关文章 1

编辑推荐

Metrics

本文评价