【正文】
全面的研究,重點(diǎn)分析了MSRM算法在圖像分割中的應(yīng)用,實(shí)現(xiàn)了高效和便捷的自適應(yīng)圖像分割,本文所做的主要工作如下:(1)總結(jié)了圖像與視頻分割技術(shù)的發(fā)展歷程,重點(diǎn)比較分析了各種類型圖像與視頻分割技術(shù)的特點(diǎn),側(cè)重閾值分割、間斷檢測(cè)、區(qū)域分割三大類分割技術(shù),深度挖掘了不同類型分割技術(shù)的優(yōu)勢(shì)和需要改進(jìn)之處。 明確了圖像預(yù)分割處理在圖像與視頻提取技術(shù)中起到的制約作用。(2)對(duì)基于區(qū)域合并的紋理圖像分割算法進(jìn)行一些研究,交待了區(qū)域的表示和相似性度量,闡述了基于最大相似度的區(qū)域合并機(jī)制,重點(diǎn)分析了區(qū)域合并過(guò)程,最后對(duì)合并算法的收斂性進(jìn)行了證明。(3)根據(jù)本文的改進(jìn)思想,設(shè)計(jì)出了一套自動(dòng)化圖像分割系統(tǒng),并驗(yàn)證了該系統(tǒng)具有較好的實(shí)時(shí)性和較高的自動(dòng)化程度。同時(shí),結(jié)合通過(guò) 圖像處理工具箱得到的仿真結(jié)果,驗(yàn)證了本文改進(jìn)思想的可行性,有效性。參考文獻(xiàn)[1] Gonzalez R C,Richard E W.Digital image processing.2nd ed.Beijing:Publishing House of Electronics Industry,2003,460521 .[2] 章毓晉.圖象分割.北京科學(xué)出版社,2001.[3] 邊肇祺,張學(xué)工等.模式識(shí)別(第二版).北京:清華大學(xué)出版社,2000.[4] K.Karu,A.K.Jain,R.M.Bolle.Is there any texture in the images.PatternRecognition,1996,29(9):1437—1446[5] Fabien S,Wojciech P.Parameter estimation in hidden fuzzy Markov random fields and image segmentation. Graphical Models and Image Processing, 1997,59(4):205220 [6] Zhang Y Y,Stephen S,Michael B.Segmentation of brain MR images using Markov random field.In:Proceedings of Medical Image Understanding and Analysis,1999,6568 [7] 葉俊勇,汪同慶,彭健等.基于小波分形的圖像分割算法.計(jì)算機(jī)科學(xué),2002,29(12):157159,149 [8] Sang H P,Dong Y,Sang U L.Color image segmentation based on 3D clustering: morphological approach.Pattern Recognition,1998,31(8):10611076 [9] Philippe S . Segmentation of digitized dermatoscopic images by twodimensional color clustering.IEEE Transaction On Medical Imaging,1999,18(2):164171[10] Enno L,Helge R.Adaptive color segmentationa parison of neural and statistical methods.IEEE Transaction On Neural Networks,1997,8(1):175185 [11] Kurugollu F,Sankur B,Harmanci A E.Color image segmentation using histogram multithresholding and fusion.Image and Vision Computing,2001,19(13):915928 [12] Wong F , Nagarajan R , Yaacob S . Fuzzy image segmentation via thresholding.Journal of the Institution of Engineer (india) Electronics and Telemunication Engineering Division,2006,86:6872 [13] Mao K Z, Peng Z, Tan P H. Supervised learningbased cell image segmentation for immunohistochemistry.IEEE Transactions on Biomedical Engineering,2006, 53(6):11531163 [14] Blinn J F.What is a pixel.IEEE Computer Graphics amp。 Applications,2005,25(5):8287[15] 陸劍鋒,林海,潘志庚.自適應(yīng)區(qū)域生長(zhǎng)算法在醫(yī)學(xué)圖像分割中的應(yīng)用.計(jì)算機(jī)輔助設(shè)計(jì)與圖形學(xué)學(xué)報(bào),[16] 張發(fā)存,趙曉紅,王忠等.區(qū)域生長(zhǎng)法圖像分割的數(shù)據(jù)并行方法研究.計(jì)算機(jī)工程,2004,30(17):1416 [17] Carevic D , Caelli T . Regionbased coding of color image using KarhunenLoeve transform.Graphics Models and Image Processing,1997,59(1):2738[18] 楊暉,曲秀杰.圖像分割方法綜述.電腦開發(fā)與應(yīng)用,2005,18(3):2123[19] Comaniciu.V.Ramesh and P.Meer.Real—time tracking of nonrigid objects using mean shift.In Proc.IEEE Conf.on Computer Vision and Pattern Recognition,Hilton Hcad,SC,olume II,June,2000。PP.1 42—1 49.[20] S.Chien.Y.Huang and L.Chen.Predictive watershed:a fast watershed algorithm for video segmentation.IEEE Transactions on circuit and systems for video technology.2003,13(5):453461.[21] Levin and A.Rav.Acha and D.Lischinski.Spectral matting.In Proceedings of the IEEE onfefence on Computer Vision and Pattem Recognition,2007.[22] Meyer and S.Beucher.Morphological segmentation.Joumal of Visual Communication and Image Representation.1990,l(1):2l46.[23] EDISON圖像分割軟件.://[24] M. Sonka, V. Hlavac, R. Boyle, Image Processing, Analysis and Computer Vision,Thomson, 2007.[25] Y. Li, J. Sun, C. Tang, H. Shum, Lazy snapping, SIGGRAPH 23 (2004) 303–308.[26] 朱付平,田捷,林瑤,葛行飛.基于Level Set方法的醫(yī)學(xué)圖像分割.軟件學(xué)報(bào).2002,(9):18661872.[27] Pen,J.Malik,Leaming a classification model for segmentation,ICCV03,v01.1,pp.10.17,Nice,2003.致 謝本論文是在我的導(dǎo)師鄭慶慶的悉心指導(dǎo)和大力支持下完成的。自撰寫論文以來(lái),鄭老師細(xì)致而具體指導(dǎo)我們每一個(gè)同學(xué),并且指導(dǎo)性的提出許多寶貴的意見與建議。鄭慶慶導(dǎo)師是位學(xué)識(shí)淵博、親切和藹的一名教師,在她這里我們不僅掌握了此次論文的相關(guān)技能和知識(shí),而且從她身上學(xué)到了很多做人的道理。同時(shí),我還要感謝我的同學(xué)鄭凱、徐海洲等在畢業(yè)設(shè)計(jì)期間給予我的大力支持與幫助。他們的睿智與熱心,給了我許多的啟發(fā)和幫助,對(duì)我的論文的順利完成作了許多的貢獻(xiàn)。在此,我表示誠(chéng)摯的謝意。特別真摯的感謝我的父母,他們的養(yǎng)育和教導(dǎo)之恩是我立身行事的根本。感謝他們一直以來(lái)無(wú)私地支持我、鼓勵(lì)我,使我能夠安心地學(xué)習(xí),學(xué)業(yè)得以順利完成。家人的支持和理解,永遠(yuǎn)是我前進(jìn)的動(dòng)力。最后感謝我的母校武漢科技大學(xué),感謝四年來(lái)的培養(yǎng),讓我各方面都有