Please use this identifier to cite or link to this item:
標題: Dog Nose Region Image Segementation
作者: 楊千瑩
Chien-Ying Yanng
關鍵字: dog nose prints
image segm
引用: [1]H. S. International. Street Dog Welfare in India. Availabble: [2]H. S. International. Street Dog Welfare in China. AvailaCable: [3]T. M. H. Society. Suffering in Slums: The global stray dog problem Availabledm.http://wwww.michiganhumane.orgg/site/News2?page=NeewsArticle&&id=11701&&new s_iv_ctrl==2261 [4]J. McGrath How Pet Microchipping Works. Available:http://sciennce.howstuffworks.comm/innovatioon/everydayy-innovationns/pet-microochip.htm [5]L. T. Dickert. Dogs Nose prints C be Used to Prove Identity, Just Like Fingerprints Availabnts.ble: e-fingerprints [6] N. Coldea 'Nose prints as a method of identification in dogs,' Veterinary Quarterly, vol. 16, pp 60-60, 19,p.994/04/01 19994. [7]H.-F. Ng, 'Automatic thresholding for defect detection,' Pate Recognition Letters, vol. 27, pp. 1644-1649, 10/15/ 2006ol.6. [8]N. Otsu, 'A threshold selection method from gray-level histograms,' IEEE Transaction Systems, Man an Cybernetics, vol. 9, pp. 62-66, 01// 1979. [9]W. M. Hafizah and E. Supriyanto 'Automatic generation of region of interest for kidney ultrasound images using texture analysis,' Int J Biol Biom Eng, vol. 6,medv pp. 26-34, 2012., [10] R. C. Gonzalez, R. E. Woods, a S. L. Eddins, Digit Image Processing Using MATLAB: Prentice-H Inc., 20Hall,003. [11] K. Pearso 'Contribon,butions to t Mathemthematical Theory of Evoolution. II. Skew Variation in Homogeeneous Maaterial,' Phiilosophical Transactio of the RoyalonsR Society of London. A, vol. 186, p 343-414, 1895.fpp. [12] P. Shanmmugavadivu and K. Balasubrammanian, 'PParticle swwarm optimized multi-objeective histoggram equallization for image enhaancement,' Optics & LaserL Technolog vol. 57, pp. 243-251 4//,p1, [13] K. S. Sim, C. P. Tso, and Y. Y. T 'RecurTan,rsive sub-immage histogrram equalization applied to gray scale images,' Paattern Recoognition Letters, vol. 28 pp. 1209-8,-1221, 7/15/ 20077. [14] K. Yeong--Taeg, 'Conntrast enhanncement usi brightne preservi bi-histoingessingogram equalizatio IEEE Transaction onConsum Electronics, vol. 43 pp. 1-8, 1997.on,'Tnsmer3,1 [15] G. Xu, J. Su, H. Pan Z. Zhang, and H. Gong, 'An Image Enhancement Method Based on Gamma Correction,' in Second International Symposium on Computational Intelligence and Design, 200 , 2009, pp. 60-63.09.p [16] T. Xiaoyang and B. Triggs, 'Enhanced Local Texture Feature Sets for Face Recognition Under Difficult Lighting Conditions,' IEEEE Transactions on Image Processing vol. 19, pp. 1635-16g,p650, 2010. [17] Z. Dongni P. Won-Ja L. Seung-Jun, A. C. Kang, an K. Sung-Jea, 'Histogram partition based gamma correction for image contrast enhancement in 2012 IEEE' 16th Interrnational Syymposium on Consumer Electronic (ISCE), 2nrcs2012, pp. 1-4. [18] S. W. Zucker and D. Terzopoulos, 'Finding structure in Co-occurrence matrices for texture analysis,' Computer Graphics and Image Processing vol. 12'rsgeng,2,286-308, 3 1980.3// [19] R. M. Harralick, K. Shanmugam and I. H. Dinstein, 'Tm,Textural Features for Image Classificattion,' IEEE Transaction Systems, Man an Cybernendetics, vol. SMMC-3,pp. 610-62 1973.21, [20] D. A. Claausi and M. E. JernigMgan, 'A fas method to determin co-occursttnerrence texture feaatures,' IEE TransactEEtions on Geeoscience an Remote SndSensing, vo 36,ol.pp. 298-30 1998.00, [21] G. Jones, P. Willett, R. C. Glen A. R. Lean,ach, and R. Taylor, 'DDevelopmen andnt validation of a gene algorith for flexnetichmxible docking,' Journa of Molealecular Biology, v 267, pp. 727-748, 4 1997.vol..4/4/ [22] D. E. Gooldberg, Genetic Algoorithms in Search, Optimization and MaOnachine Learning: Addison-WWesley Longgman Publisshing Co., Inc., 1989. [23] Y. Yun, 'HHybrid gene algorith with adaetichmaptive local search scheeme,' Compputers & Industri Engineeialering, vol. 5 1, pp. 128-1 9// 200141,06. [24] U. Maulik 'Medica Image Sk,alSegmentatio Using Genetic AlgonGgorithms,' IEEE Transactio on Infoonsormation Teechnology in Biomedicicine, vol. 1 pp. 16613,6-173, 2009. [25] R. M. Harralick and L. G. Shapi Comput and Rob Vision: Addison-WLiro,terbotWesley Longman Publishing Co., Inc., 11992. [26] A. Rosenffeld, 'Connnectivity in Digital Picctures,' J. ACM, vol. 1 pp. 146A17,6-160, 1970. [27] Y. Yang, Y Peng, and Z. Liu, 'A Fast AlgorY.dArithm for YCbCr to RG ConversYGBsion,' IEEE Trannsactions on Consumer Electronic vol. 53, pp. 1490-14nrcs,p493, 2007. [28] G. H. Jooblove and D. Greendnberg, 'Coolor spaces for compsputer graphhics,' SIGGRAP Comput. Graph., vo 12, pp. 20-25, 1978.PHol. [29] M. W. Schhwarz, W. B. Cowan, a J. C. Beatty, 'An experimenta comparis ofBandealson RGB, YIQ LAB, HS and oppQ,SV,ponent color models,' ACM Trans Graph., vol. 6,rAs.v pp. 123-15 1987.58, [30] D. M. Poowers, 'Evvaluation: ffrom precission, recall and F-meleasure to ROC,R informednness, markeddness and ccorrelation,' 2011.' [31] Y. J. Huan R. Powers, and G. T. Monteling,ione, 'Prote NMR ReinRecall, Precision, and F-meeasure Scor (RPF S cores):  Strresructure Quaality Assesssment Meaasures Based on Informatio Retrieva Statistics, Journal of the Ameonal,'erican Chemical Society, vo 127, pp. 1665-1674 , 2005/02/0 2005.ol.01
摘要: Because of human abandonment and the lack of the concepts of neutering ,the problems of stray dogs are increasing. Stray dogs also derive many social problems, like infectious diseases, bite, influence on quality of life, etc. To deal with these problems, every country rounded up stray dogs and treated them in inhumane way before. And now, they use pet microchips pet collars, animal shelter, etc. to manage stray dogs. In this thesis, we propose a new management for stray dog and pets with computer technology. Because of the unique do nose prints we want to develope a system that can recognize d nose dog prints. The system consists of two parts: image segmentation and nose prints recognition In this thesis, we focus on the former part, segmentation of dog nose. First of all the proposed method uses image processing technology to enhance contrast of images. Then the appropriate features are derived from Gray Level Co-occurrence Matrix. According to those feature the background and nose region be identified from dog no images. Finally, Otsu's method is used to segment dog noses. In experimental results, the average precision, average recall and average F-measure are larger than 90%. And the average accuracy is higher than 88%.
由於人們棄養寵物及對結紮、節育觀念之缺乏,流浪狗的問題日益增加,並產生生許多社會問問題,如疾病傳染、打鬥及影響生活品質等。面對流浪狗之問題,各國從以前利用圍捕等不不人道的對待待方式,一直到現今常用之寵物晶片、寵物項圈、收容所等,加以規範管理流浪狗。因此,,我們提出一種利用電腦科技管管理流浪狗及寵物之方式,利用狗鼻紋之唯一性,欲欲發展一套狗鼻紋辨識系統,而辨識系統主要可分為兩個部分分:影像切割及鼻紋辨識,本篇主要研究於前者,將鼻子部位從狗鼻子影像中切割割出來。 首先,本篇利用影像處理技術將影像對比增強,接著使用 Gray Level Co-ooccurrence MMatrix 找出適合之特徵,將影像中背景與鼻子區域予以分辨。最後,利用 Otsu's 二用二值化方法將影像切割割出來。實驗結果顯示,本篇平均精確度、平均召回率及平均 F-measure 皆有達到 9皆90%以上,而平均準確性也達到88%以上。
其他識別: U0005-2404201514592000
文章公開時間: 2018-05-11
Appears in Collections:資訊管理學系



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.