超像素分割算法研究综述_王春瑶

所需积分/C币:11 2018-09-10 17:08:00 849KB PDF
31
收藏 收藏
举报

超像素能够捕获图像冗余信息,降低后续处理任务复杂度,已受到了国内外研究者的日益关注。首先 分析了超像素分割领域的发展现状,以基于图论的方法和基于梯度下降的方法为视角,对现有超像素分割方法 进行归纳和论述。在此基础上,就目前常用的超像素分割算法进行了实验对比,分析各自的优势和不足。最后, 对超像素分割技术的最新应用进行了介绍和展望
31 H′A+ABA A∈EH′A 入≥0 X-X|t∈T 3 6 X =limAX,IX,Ix, 7 G= vA ∑1 Pii A log p;A T 7 superpixel lattice ifi≠ i and e;;∈A A A ∑ if i=j 009 b ●● (熵率=081滴率=043()嫡率=064)率=061 BA=HA-NA=-∑ 10 o Lattice-cut A IS. (a) (b) c) (d) a)平衡值=-1.00 (b平衡值=-1.19 21994-2018ChinaAcademicJournalElectronicPublishingHouse.Allrightsreservedhttp://www.cnki.net 9 Turbopixcls 8 a b Mean-shift Fukunaga 1995 Cheng Mean-shift Comaniciu Mean-shift Mean-shift (a放置种子点(b第次演化(o)更新骨架 7 Mean-shift (更新速度()度才展 8 Turbopixels ean Turbopixel gla shift Turbopixel P Ncut 4 SLIC Mean-shift Achanta CIELAB X Mean-shift Ncut Mean-shift N/K N/K 2.3 Turl 2009 Levinshtein ID 3×3 21994-2018ChinaAcademicJournalElectronicPublishingHouse.Allrightsreservedhttp://www.cnki.net 10· 31 3.00GB Intel@ Core 2 yk-; D i =die 21×481 1.16 SLIC 1.23 0.36 S×2S 2003 en 9 (a)在整张图像中搜索(bSC在限定区域内搜索 CPU NVIDIA CUDA SLIC Ncut 10~20 SLIC SLIC SLIC 15 supervoxels 3D Berkeley benchmark Kohli entropy rate El°、 Ncut-based Ncut-o18 perpixel lattice SL .\Turbopixels TP SLIC 12 321×481 00 10 ER 1.16s。 Neul- (始EM图像h像素 (c)超像素图(d)最后分割结果 F, M 321 M 48I 2 min M TP SLIC 12 因图A Ncut-B TP 21994-2018ChinaAcademicJournalElectronicPublishingHouse.Allrightsreservedhttp://www.cnki.net on and Pattern Reeognition. Washington DC IEEE Computer Socie ty1997731-737 SHI Jiani+o MALIK J. Normalized cuts and image segmentation J IEEE Trans on Pattern Analysis and Machine Intelligence 5 MOORE A PRINCE S WARRELL J et al. Superpixel lattices C//Proc of IEEE Conference on Computer Vision and Pattern Ree 200818 6 LIU Ming-yu TUZEL O RAMALINGAM S et al. Entropy rate su perpixel segmentation C //Proc of IEEE Conference on Computer Vision and pattern recognition 2011 2097-2104 7 VINCENT'L SOILLE P. Watersheds in digital spaces an efficient algorithm hased (n immersion simulations J.IEEE Trans on Pat- tern Analysis and machine Intellig 1991136583598 8 COMANICIU D MEER P. Mean shift a robust approach toward fea ture alysis J. IEEE Trans on Pattern Analysis and Ma chine Intelligence 2002 24 5 603-619 ia)bird #7 (h bird #11 (e)bird #19 9 VEDALDI A SOATTO s. Quick shift and kenel methods for mode seeking M / Computer Vision. Berlin Springer-Verlag 2008 705-718 10 LEVINSHTEIV A STERE A KUTULAKOS Kn e u. turbopi ()lemming#867 (e)lemming 41055 (flemming #1 102 xels fast superpixels using geometric flows J. IEEE Trans on Pattern Analysis and Machine Intelligence 2009 31 12 2290-2297. 1 ACHANTA R SHAJI A SMITH K et al. slic superpixels EPFL 149300R Ecole Polytechnique Federal de L. ausssanne (h)transformer #52 transformer #124 EPFL 2010 13 12 ACHANTA R SHAJI A SMITH K et al. SLIC superpixels com bared to state-of-the-art superpixel methods ]. IEEE Trans on Pattern Analysis and Machine Intelligence 2012 34 Il SIFT 2274-2282. Tighe 13 LEVINSHTEIN A SMINCHISESCH C DICKINSON S. Multiscale symmetrie part delection and grouping J. International Journal of 14 LUCCHI A SMITH K ACHANTA R et aL. A fully automated ap Graz-02 PASCAL VOC 2007 proach to segmentation of irregularly shaped cellular structures in EM images C /Medical Image Computing and Computer-Assisled Inter- vention. Berlin Springer-Verlag 2010 463-471 15 LUCCIII A SMITII K ACIIANTA R et al. Supervoxel-ased seg mentation of mitochondria in EM image stacks with leamed shape fea tures J. IEEE Trans on Medical Imaging 2012 31 2 474 16 KOHLI P TORR P H S. Robust higher order potentials for enforcing label consistency J. International Journal of Computer Vision 2009823302-324. 17 MORI G rEn Xiao -feng EFROS AA et al. Recovering human body configurations Combining spymenlalion and rer'ngnilion C // Proe of IEEE Computer Society Conference on Computer Vision and Patlern Recognition. Washington DC IEEE Computer Society 2004 326-333 1 REN Xiao-feng MALIK J. Learning a classification model for seg 18 MORI G. Guiding model search using segmentation C // Proe of the mentation C //Proc of the 9th IEEE International Conference on 10th IEEF Inlernational conference on Con Computer Vision. Washington DG IEEE Computer Society 2003 10-1 DC IEEE Computer Society 2005 1417-1423 2 FELZENSWALB P F HUTTENLOCHER D P. Efficient graph-based 19 PANTOFARU C SCHMID C HEBERT M. Object recognition by in Image segmentation J International Journal of Computer tegrating multiple image segmentations C !/Proc of the 10th Europe Vision2004592167-181 an Conference on Computer Vision. Berlin Springer-Verlag 2008 3 SHI Jian -Do MALIK J. Normalized cuts and image segmentation 481-494 C//Proc of IEEE Computer Society Conference on Computer Vi- 20 FULKERSON B VEDALDI A SOATTO S. Class segmentation and 21994-2018ChinaAcademicJournalElectronicPublishingHouse.alLrightsreservedhttp:/www.cnki.net 12 31 object localization with superpixel neighborhoods C //Proe of the 31 CHENG Yi-zong. Mean shift mode seeking and clustering J 12th International Conference on Computer Vision. Washington DC IEEE Trans on Patten Analysis and Machine Intelligence IEEE Cormputer Society 2009 670-677 1995178790-799 WANG Shu LU Hu-chuan YANG Fan et al. Superpixel tracking 32 WANG Jue thieSSon B XU Ying -ging et al. Image and video se C //Proc of IEeE International Conference on Computer Vision mentation by anisotropic kernel mean shift C //Proc of European 20111323-1330 Conference on Computer Vision Berlin Springer-Verlag 2004 238-249 22 ZHOU Xue LI Xi CHiN T J et al. Superpixel-driven level 33 TAO Wen-bing JIN Hai ZHANG Yi-ming. Color image segmenta tracking C //Pre the 19th ieee International Conference tion based on mean shift and normalized cuts I. EEE Trans on Image Processing. 2012 409-41 Systems Man and Cybernetics Part B Cybernetics 2007 3 LIU Li-wei XING Jun iang AI Hai zhou et al. Semantic super 3751382-1389 pixel based vehicle tracking C //Proc of the 21 st International Con- 34 XIANG Shi-ming PAN Chun -hong NIE Feiping et al. Turbopixel fe-rence on Pattern Recognition. Berlin Springer-Verlag 2012 2222 segmentation using Eigen-images J.IEEE Trans on Image Pro- 22 cessing 2010 19 11 24 WANG Wei-jun NEVATIA R. Robust object tracking using constel- 35 CIGLA C ALATAN A A. Efficient graph-based image segmentation lation model with superpixel C //Proc of the 1 1th Asian Conferenee via speeded-up turbo pixels C //Proc of the 17th IEEE International g.20103013-3016 25 HOIEM D EFROS AA HEBERT M. Automatic photo pop-up J 36 REN C Y REID I. gSLIC a real-time implementation of SLIC. super ACM Trans on Graphics 2005 24 3 577-584 pixel segmentation R Qx ng Univer 26 MAJI S VISIINOINK MALIK J. Biased normalized cuts C i sity of Oxford 2011 Proc of IEEE Conference on Computer Vision and Pattern Recogni 37 LUCCHI A LI Yun-peng SMITH K et al. Structured image seg tion. Washington DC IEEE Computer Society 201 1 2057-2064 mentation using kernelized features C //Proc of the 12th European 27 MOORE A PRINCE S. Scene shape prioRs fur superpixel segmenta- International Conference on Computer Vision. Berlin Springer-verlag tion C //Proc of the 12th IEEE International Conference on Compu- 2012400-413 ter Vision. Washington DC IEEE Compuler Society 2005771-778 38 BOYKOV YY JULLY M P. Interactive graph cuts for optimal 28 MOORF A PRINCE. SJ D WARRFLL J. lattice (ul onst rutting segmentation of objects in N-D)images C / superpixels using laver constraints C //Proc of IEEE Conference on Proe of the &th IEEE International Conference on Computer Vision Computer Vision and Pattern Recognition. 2010 2117-2124 200l105-112 29 DIG ABEL. H LANTUEJOUL C. Iterative algorithms C//Pr 39 GU Guang Hula LI Feng 4: i ZHAOY At a/. Scene classificat ion the 2nd European Symposium on Quantitative Analysis of M based on spatial pyramid representation by superpixel lattices and con tures in Material Science Biology and Medicine. Sturrgart Riederer- textual visual features Optical Engineering 2012 51 1 Verlag 1978 017201-1-017201-8 30 FUKUNAGA K HOSTETLER L. The estimation of the gradient of a 40 TICHE J LAZEBNIK S. Superparsing scalable nonparametric image density function with applications in pattern recognition J. IEEE parsing with superpixels C//Proc of the 1 lth European Computer Trans on Information Theory 1975 21 1 32-40 Vision. Berlin Springer-Verlag 2010 352-365 22 MYERS M YOUNDT S. An introduction to hardware-assisted virtual 14 DESNOS A FILIOL E LEFOU I. Detection of an HVM rootkit aka machine HVM rootkits R. S.l. WhitePapers DB. 2007 BluePilllike J. Computer Virology 2011 7 1 23-49 23 LI He-shuai ZHU Jun-hu ZHOU Tian-yang et al. A new mecha- 5 KYTE I ZAVARSKY P LINDSKOC D. Enhanced side-channel a- nism for preventing HVM-aware malware C //Proc of the 3 rd Inter- nalysis method to detect hard ware virtualization based rootkits c / national Conference on Communicution Software and Networks. 2011 Proc of World Congress on Internet Security. 2012 192-201 163-167 16 RUTKOWSKA Red Pill or how to detect VMM using almost one 24 SHEN I LIPASTI M. Modern processor design fundamentals of su CpuinstructionEb/oL.2004-11-17.http://www.securite- Derscalar processors M. S.I. MeGraw-Hill 2005 am. com/securityreviews/6Z0oH2UBQS html 25 MILENKOVIC M MILENKOVIC A KULICK J. Demystifying Intel 7 RLTKOWSKA. Blue Pill detection EB/OL. 2006-03-12. ht branchpredictorseb/ol.http!/www.ece.wiseedu/-wddd/ tp //theinvisiblethings. blogspot. com/2006/08/blue -pill-etection ali mllenKovIC html 26 FERRIE P. Attacks on more virtual machine emulators EB/OL 18 QLIST D SMITH V. Detecting the presence of virtual machines using 2007-08-31.http://pferrie.tripodcom/papers/attacks2.pdf the local data table EB/OL. htlp //index of es/Misc/vm. pdf 27 BULYGIN Y SAMYDE D. Chipset based approach to delect virtuali 19 FERRIE. P. Allac:ks on virtual machine emulators FB/OL.. htlp vation malware R. S.I. InTel Corporation 2008 www.symanteccom/avcenter/reference/virtualMachineThreats 28 GARFINKEL T PFAFF B CHOWJ et al. Terra a virtual machine Ised platform for trusted computing C //Proc of the 19th ACM 20 ADAMS K. BluePill detection in two easy steps EB/OL. 2007 Symposium on Operating Systems Principles. New York ACM Press 07-02.http://x86vmm.blogspot.com/2007/u7/bluepill-detection- 2003103-206 in-two-easy-steps% o. html 29 New Bluepill 21VirtualizationtechnologyforAmdarchitectureEb/ol.http:// 0l155 download. microsoft. com/download /9/8/f/9813fe47-Ifc3-4e74-92a3 30 WMware. Timekeeping in VMware virtual machines EB/OL.ht- 088782200fe7/twar05014 winhec05 pdf tp//www.vmwarecom/pdf/vmwaretimekeeping.pdf 21994-2018ChinaAcademicJournalElectronicPublishingHouse.alLrightsreservedhttp:/www.cnki.net

...展开详情
立即下载 身份认证VIP会员低至7折
一个资源只可评论一次,评论内容不能少于5个字
您会向同学/朋友/同事推荐我们的CSDN下载吗?
谢谢参与!您的真实评价是我们改进的动力~
关注 私信
上传资源赚钱or赚积分
最新推荐
超像素分割算法研究综述_王春瑶 11积分/C币 立即下载
1/0