跳到主要內容

臺灣博碩士論文加值系統

(3.236.84.188) 您好!臺灣時間:2021/08/06 12:06
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果 :::

詳目顯示

: 
twitterline
研究生:劉政憲
研究生(外文):Liu, Chenghsien
論文名稱:應用於肝臟超音波影像分析之影像切割及樣本擷取
論文名稱(外文):Segmentation And ROI Extraction For Liver Ultrasound Image Analysis
指導教授:郭忠民郭忠民引用關係楊乃中
指導教授(外文):Kuo, ChungmingYang, Naichung
口試委員:郭忠民楊乃中謝朝和丁慧枝
口試委員(外文):Kuo, ChungmingYang, NaichungHsieh,ChaurhehDing, Hueischjy
口試日期:2012-07-05
學位類別:碩士
校院名稱:義守大學
系所名稱:資訊工程學系
學門:工程學門
學類:電資工程學類
論文種類:學術論文
論文出版年:2012
畢業學年度:100
語文別:中文
論文頁數:78
中文關鍵詞:超音波影像影像分析
外文關鍵詞:Ultrasound ImagesImage AnalysisOtsuLBP
相關次數:
  • 被引用被引用:0
  • 點閱點閱:1354
  • 評分評分:
  • 下載下載:45
  • 收藏至我的研究室書目清單書目收藏:0
在肝臟超音波影像做切割是非常困難的,因為超音波影像中包含了太多的雜訊及衰減的部分。肝臟紋理的分佈上相當的均勻,很難基於紋理分佈的不同來進行切割。超音波影像通常是用來評估肝臟的疾病例如肝腫瘤、肝癌最常使用的儀器,而超音波影像診斷往往著重於操作者的經驗,很難提出較為客觀的臨床診斷依據。
本文中將要探討有效的肝臟超音波影像切割以及分析肝臟纖維化的特徵,在切割上因為超音波肝臟影像中包含太多雜訊干擾,所以我們將先對影像做過濾部分雜訊的動作,再利用Otsu的方法自動找出適應性的門檻值,並且加上膨脹與侵蝕的方法將我們所感興趣的肝臟部分切割出來做影像分析。在分析上將以像素灰階值做局部二位元圖形LBP(local binary pattern)運算來進行分析紋理特徵,並且建立特徵資料庫後再計算特徵資料的主要分佈,最後再利用未建立資料庫的肝臟影像做影像分析並且利用分析數據來判斷肝臟的健康與否。
經過實驗結果證明,本論文所提出的方法能夠有效的切割出肝臟的ROI(Regions Of Interest)。透過實驗結果的驗證,我們提取的ROI影像在實驗中有很好的檢索率,也代表著我們的ROI影像是可靠的。
Segmentation of ultrasound liver-images is a challenge because these images contain strong speckle noise and attenuated artifacts. It is difficult to segment livers from ultrasound images by texture properties because the homogeneity of the liver-images.
The ultrasound liver-images are usually used to examine liver diseases such as liver tumor and liver cancer. The diagnosis of the ultrasound liver-images is highly dependent on operator's experience; it's difficult to propose an objective evaluation for Clinical diagnosis.
This thesis aims at the liver segmentation from ultrasound liver-images, and fibrosis analysis of the segmented livers. We first filter the speckle noise in image, and then use Otsu’s method to find the adaptive threshold. Next, we use dilation and erosion method to segment livers from ultrasound liver-images. We propose an improvement for LBP (local binary pattern) method, which is used to analyze texture features.
Experimental results show that our proposed method provides effective ROI segmentation. Furthermore, the segmented results can be identified with a high retrieval rate.
摘要I
Abstract III
謝誌V
圖目錄VIII
一、緒論1
1.1 研究動機與問題描述1
1.2 論文架構5
二、文獻回顧與探討6
2.1 邊緣偵測技術6
2.2 區域生成技術9
2.3 紋理切割技術10
2.4 二值化切割法12
三、超音波影像切割及取樣13
3.1 前處理18
3.2 超音波影像切割22
3.3 肝臟主體匹配28
3.3.1 二值化28
3.3.2 肝臟偵測29
3.3.3 Closing 運算31
3.3.4 ROI的擷取34
四、超音波影像特徵分析及檢索36
4.1 LBP特徵擷取的方法改良37
4.2 主要分佈的擷取41
五、肝臟切割與檢索實驗結果42
5.1 肝臟超音波影像切割44
5.2 肝臟超音波影像特徵檢索50
六、結論與未來發展61
參考文獻62
圖1-1 肝臟分佈示意圖3
圖2.1-1 Sobel 濾波器7
圖2.1-2 Sobel Filter影像轉換結果8
圖2.1-2 Sobel Filter影像轉換結果8
圖2.2-1 像素種子區域生長法10
圖2.3-1 演算法分割流程11
圖3-1 肝臟分佈示意圖15
圖3-2 正常肝與纖維肝比較圖15
圖3-3 切割取樣流程圖16
圖3.1-1 肝臟固定範圍切割示意圖18
圖3.1-2正規化範例21
圖3.2-1 探頭固定所及的區域22
圖3.2-2 切割影像示意圖24
圖3.2-2 侵蝕與膨脹示意圖26
圖3.2-3 切割示意圖27
圖3.3-1 肝臟主體匹配流程圖28
圖3.3.2-1 肝臟區域萃取示意圖30
圖3.3.3-1 肝臟影像萃取示意圖32
圖3.3.3-2 切割、匹配示意圖33
圖 3.3.4-1 小區塊切割範例圖35
圖4-1 特徵分析檢索流程圖36
圖5.1 影像切割介面43
圖5.2 影像分析介面43
圖5.1-2 切割實驗結果影像46
圖5.1-3 切割實驗結果影像48
圖5.1-4 切割實驗結果影像49
圖5.2-1 3×3 CO-LDP 51
圖5.2-2 3×3 CO-LDP 檢索圖52
圖5.2-3 檢驗結果與罹病狀況表示圖52
圖5.2-4 纖維肝臟影像1,ROI張數 25張,區塊大小75×75 54
圖5.2-5 纖維肝臟影像2,ROI張數18 張,區塊大小75×75 54
圖5.2-6 正常肝臟影像1,ROI張數 5張,區塊大小75×75 55
圖5.2-7 正常肝臟影像2,ROI張數 19張區塊大小75×75 55
圖5.2-8 纖維肝臟影像1,ROI張數 50張區塊大小50×50 56
圖5.2-9 纖維肝臟影像2,ROI張數 47張區塊大小50×50 57
圖5.2-10 正常肝臟影像1,ROI張數 37張區塊大小50×50 57
圖5.2-11 正常肝臟影像2,ROI張數 46張區塊大小50×50 58
圖5.2-12 CO-LDP區塊大小75×75 59
圖5.2-13 CO-FLDP區塊大小75×75 59
圖5.2-14 CO-LDP區塊大小50×50 60
圖5.2-15 CO-FLDP區塊大小50×50 60
[1]Bleck JS, Ranft U, Gebel M, et al. Random field models in the textural analysis of ultrasound images of the liver. IEEE Trans Med Imaging 1996; 15:796–801
[2]Cao, G. T., P. F. Shi, et al. (2005). "Liver fibrosis identification based on ultrasound images captured under varied imaging protocols." J Zhejiang Univ Sci B 6(11): 1107-14.
[3]C. M. Wu, Y. C. Chen and K. S. Hsieh, “Texture features for classification of ultrasonic liver images,” IEEE Tans. Med.Imaging, Vol. 11,NO. 2,pp. 141-152, 1992.
[4]Cramblitt RM, Parker KJ. Generation of non- Rayleigh speckle distribution using marked regularity models. IEEE Trans Ultrason Ferroelectr Freq Control 1999; 46:867–874
[5]D.C. He, L. Wang and J. Guibert, “Texture discrimination based on an optimal utilization of texture features,” Pattern Recognition, Vol. 24, No. 2, pp. 141-146, 1988.
[6]D. K. Iakovidis, E. G. Keramidas and D. Maroulis, “Fuzzy Local Binary Patterns for Ultrasound Texture Characterization,” ICIAR Vol. 5112, pp.750-759, 2008.
[7]Fujii Y, Taniguchi N, Y. Clinical application of a new method that segments the region of interest into multiple layers for RF amplitude histogram analysis in the cirrhotic liver [in Japanese]. J Med Ultrason 2001; 28:25–33
[8]H. Toyoda, T. Kumada, N. Kamiyama, et al, “B-mode ultrasound with algorithm based on statistical analysis of signals: evaluation of liver fibrosis in patients with chronic hepatitis C ,” AJR Am J Roentgenol, Vol. 193, No. 4, pp. 1037-1043, Oct 2009.
[9]J. Serra, Image Analysis and Mathematical Morphology. London:Academic, 1982.
[10]Jiankang Wang and Xiaobo Li, “A System for Segmenting Ultrasound Images,” Pattern Recognition proceedings 14th international conference, vol. 1, pp. 456-461, 1998.
[11]Khanh Vu, Kien A. Hua and Duc A. Tran, “An Efficient Core-Area Detection Algorithm for Fast NoiseFree Image Query Processing,” In Proc. of The 16th ACM-SIGAPP Annual Symposium on Applied Computing, pp. 258-263, Mar. 2001.
[12]N. Hiransakolwong, , H. Kien. A., K. Vu, and P. S. Windyga, (2003) Segmentation of ultrasound liver images: An automatic approach. IEEE Multimedia and Expo, 2003 ICME 03. Proceedings, 2003 International Conference, 1, 573 -576
[13]N. Hiransakolwong, K.A. Hua, K. Vu, P.S. Windyga, “Segmentation of ultrasound liver images: An automatic approach,” in: IEEE Multimedia and Expo, 2003 ICME Proceedings, 2003 International Conference, vol. 1, 6–9 July, 2003, pp. 573–576
[14]N. Otsu,“A threshold selection method from gray-level histograms”, IEEE Trans. on System, Man and Cybernetics, vol. SMC-9, no. 1, pp. 62-66, 1979.
[15]Piotr S. Windyga, “Fast Impulsive Noise Removal,” IEEE Trans. Image Processing, vol. 10, no. 1, pp. 173- 179, 2001.
[16]R. M. Haralick, “Statistical and structural approaches to texture,” Proc. IEEE, Vol. 67, No. 5, pp. 786-804, 1979.
[17]R. M. Haralick, K. Shanmugam and I. Dinstein, “Texture features for image classification,” IEEE Transaction on system,mam,and cybernetics,Vol. SMC-3, NO.6,November 1973.
[18]T. Ojala, M. Pietikäinen, and David Harwood, “A comparative study of texture measures with classification based on feature distributions,” Pattern Recognition, Vol. 29, No. 1, pp. 51-59, 1996.
[19]Xiaohui Hao, Charles Bruce, Cristina Pislaru and James F. Greenleaf, “A Novel Region Growing Method for Segmenting Ultrasound Images,” IEEE Ultrasonics Symposium, vol. 2, pp. 1717-1720, 2000.
[20]Yamada H, Ebara M, Yamaguchi T, et al. A pilot approach for quantitative assessment of liver fibrosis using ultrasound: preliminary results in 79 cases. J Hepatol 2006; 44:68–75
[21]Z. Xu and Xiaoming Liu, “Bilinear Interpolation Dynamic Histogram Equalization for Fog-degraded Image Enhancement,” Journal of Information & Computational Science 7: 8, pp. 1727-1732, 2010.
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top