|本期目录/Table of Contents|

[1]崔 哲,孟凡荣,姚 睿*,等.CUDA并行计算下基于扩展SURF的多摄像机[J].南京大学学报(自然科学),2016,52(4):627.[doi:10.13232/j.cnki.jnju.2016.04.007]
 Cui Zhe,Meng Fanrong,Yao Rui*,et al.Multi­video fusion with extended SURF based on CUDA parallel computing framework[J].Journal of Nanjing University(Natural Sciences),2016,52(4):627.[doi:10.13232/j.cnki.jnju.2016.04.007]
点击复制

CUDA并行计算下基于扩展SURF的多摄像机()
     

《南京大学学报(自然科学)》[ISSN:0469-5097/CN:32-1169/N]

卷:
52
期数:
2016年第4期
页码:
627
栏目:
出版日期:
2016-08-01

文章信息/Info

Title:
Multi­video fusion with extended SURF based on CUDA parallel computing framework
作者:
崔 哲1孟凡荣1姚 睿1*石记红12
1.中国矿业大学计算机学院,徐州,221116;2.郑州煤炭工业(集团)有限责任公司,郑州,450042
Author(s):
Cui Zhe1Meng Fanrong1Yao Rui1*Shi Jihong12
1.School of Computer Science,China University of Mining and Technology,Xuzhou,221116,China;2.Zhengzhou Coal Industry(Group)Co.,Ltd.,Zhengzhou,450042,China
关键词:
视频融合speeded up robust features(SURF)k-d树Compute Unified Device Architecture(CUDA)
Keywords:
video fusionSURF(speeded up robust features)k-d treeCUDA(Compute Unified Device Architecture)
分类号:
TP317
DOI:
10.13232/j.cnki.jnju.2016.04.007
文献标志码:
A
摘要:
在多摄像机视频融合过程中,需要对多个摄像机获取的视频中的每一帧图像进行大量诸如特征提取、图像配准、图像融合等高复杂度的计算,占用大量的运算时间,这对视频融合的实时性要求是一个很大的挑战.基于CUDA(Compute Unified Device Architecture)并行计算框架,提出了一种快速、可靠的多摄像头视频融合方法,该方法首先利用基于局部环形扩展及颜色描述子的SURF(speeded up robust features)特征提取方法提取图像特征点;其次采用基于分块相似性度量的k-d树(k-维树)多图像自动特征匹配算法进行图像与特征点的匹配;然后使用RANSAC(Random Sample Consensus)算法计算变换矩阵;最后使用多频率融合算法进行多摄像机视频融合,得到流畅的大视场视频.整个多视频融合过程使用CUDA进行并行加速,并在多个不同场景与摄像机数量下的实验验证了本文算法的实时性与有效性.
Abstract:
In the process of multi­camera fusion,it is necessary for each frame image acquired by multiple cameras to carry out a large number of image processing operations,such as feature extraction,image registration and image fusion.These operations take up a lot of computing time and are difficult to meet the requirements of real­time video fusion.This paper proposes a fast and reliable fusion method with multi­video camera based on CUDA(Compute Unified Device Architecture)parallel computing framework.This method firstly uses the extracting method of characteristics of SURF(speeded up robust features)which is based on partial annular extension and color descriptors to extracted image feature points;secondly,based on the block similarity measure tree,uses multi­image auto matic feature matching algorithm tomatch the images and the feature points;and then uses the RANSAC(Random Sample Consensus)algorithm to calculate the transformation matrix.Finally,multi­frequency fusion algorithm is used for multi­camera fusion.At the same time,the whole multi­video fusion process uses CUDA to carry on the parallel acceleration.The real­time and the effectiveness of the proposed algorithm are verified by experiments with a number of different scenarios and cameras.

参考文献/References:

[1] 苗立刚.视频监控中的图像融合与合成算法研究.仪器仪表学报,2009,30(4):857-861.(Miao L G.Image mosaicing and compositing algorithm for video surveillance.Chinese Journal of Scientific Instrument,2009,30(4):857-861.) [2] Zitová B,Flusser J.Image registration methods:A survey.Image & Vision Computing,2003,21(11):977-1000. [3] Lowe D G.Distinctive image features from scale­invariant keypoints.International Journal of Computer Vision,2004,60(2):91-110. [4] Ke Y,Sukthankar R.PCA­SIFT:A more distinctive representation for local image descriptors.IEEE Computer Society,2004:506-513. [5] Krystian M,Cordelia S.A performance evaluation of local descriptors.IEEE Transactions on Pattern Analysis & Machine Intelligence,2005,27(10):1615-1630. [6] 朱 进,丁亚洲,肖雄武等.基于SIFT改进算法的大幅面无人机影像特征匹配方法.计算机应用研究,2015,32(10):3156-3159.(Zhu J,Ding Y Z,Xiao X W,et al.Feature detection of large format UAV images based on improved SIFT algorithm.Application Research of Computers,2015,32(10):3156-3159.) [7] Bay H,Tuytelaars T,Gool L V.SURF:Speeded up robust features.Computer Vision & Image Understanding,2006,110(3):404-417. [8] Valgren C,Lilienthal A J.SIFT,SURF and Seasons:Appearance­based long­term localization in outdoor environments.Robotics & Autonomous Systems,2010,58(2):149-156. [9] 雷 飞,王文学,王雪丽等.基于改进SURF的实时视频拼接方法.计算机技术与发展,2015(3):32-35.(Lei F,Wang W X,Wang X L,et al.Real­time video stitching method based on improved SURF.Computer Technology and Development,2015(3):32-35.) [10] 卜 珂.基于SURF的图像配准与拼接技术研究.硕士学位论文.大连:大连理工大学,2009.(Bu K.Research on image registration and mosaic based on SURF.Master Dissertation.Dalian:Dalian University of Technology,2009.) [11] 熊云艳,毛宜军,闵华清.有序的KD­tree在图像特征匹配上的应用.化工自动化及仪表,2011,37(10):84-87.(Xiong Y Y,Mao Y J,Min Q H.Application of the ordered KD­tree on the image features matching.Control and Instruments in Chemical Industry,2011,37(10):84-87.) [12] 杜振鹏,李德华.基于KD­Tree搜索和SURF特征的图像匹配算法研究.计算机与数字工程,2012,40(2):96-98.(Du Z P,Li D H.Image matching algorithm research based on KD­tree search and SURF features.Computer and Digital Engineering,2012,40(2):96-98.) [13] Ryoo S,Rodrigues C I,Baghsorkhi S S,et al.Optimization principles and application performance evaluation of a multithreaded GPU using CUDA.In:Proceedings of the 13th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming.Salt Lake City,UT,USA:ACM,2008:73-82. [14] Huang Y,Liu J,Tu M,et al.Research on CUDA­based SIFT registration of SAR image.In:2011 the 4th International Symposium on Parallel Architectures,Algorithms and Programming.IEEE Computer Society,2011:100-104. [15] 黄成满.基于CUDA并行计算的无人机遥感图像快速拼接.硕士学位论文.成都:电子科技大学,2011.(Huang M C.Fast image mosaic of UAV remote sensing image based on CUDA parallel computing.Master Dissertation.Chengdu:University of Electronic Science and Technology of China,2011.) [16] 刘有科,高 珏,谭 松等.一种基于CUDA的快速宽视频拼接的方法.计算机技术与发展,2015(1):15-18.(Liu Y K,Gao Y,Tan S,et al.A fast wide video stitching method based on CUDA.Computer Technology and Development,2015(1):15-18.) [17] Bentley J L.K-d trees for semi­dynamic point sets.In:Proceeding of the 6th Annual(SCG’90).Symposium on Computational Geometry,1990:187-197.

相似文献/References:

备注/Memo

备注/Memo:
基金项目:国家自然科学基金(61402483,61572505,U1261201),中国博士后基金(2014M551696),中央高校基本科研业务费专项资金(2013XK10),江苏省产学研前瞻性项目(BY2015023-05) 收稿日期:2016-03-16 *通讯联系人,E­mail:ruiyao@cumt.edu.cn
更新日期/Last Update: 2016-07-23