跳到主要內容

臺灣博碩士論文加值系統

(54.172.135.8) 您好!臺灣時間:2022/01/18 14:36
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

我願授權國圖
: 
twitterline
研究生:陳淑怡
研究生(外文):Shu-Yi Chen
論文名稱:靜態影像脈衝式雜訊去除方法之研究
論文名稱(外文):A study on the method of impulsive noise removal for still images
指導教授:蕭如淵蕭如淵引用關係
指導教授(外文):Ju-Yuan Hsiao
學位類別:碩士
校院名稱:國立彰化師範大學
系所名稱:資訊管理學系
學門:電算機學門
學類:電算機一般學類
論文種類:學術論文
論文出版年:2002
畢業學年度:90
語文別:中文
論文頁數:65
中文關鍵詞:impulsive noisestandard median filtercenter weighted median filter
外文關鍵詞:脈衝式雜訊標準中位數濾波器中間權重式中位數濾波器
相關次數:
  • 被引用被引用:1
  • 點閱點閱:149
  • 評分評分:
  • 下載下載:0
  • 收藏至我的研究室書目清單書目收藏:1
在本論文中,我們針對靜態影中的脈衝式雜訊提出了三種消除的方法。在第一個方法中,我們將掃描視窗內的中間值和其周圍的值做排序,並根據排序後的順序來決定其是否為雜訊。若其順序落在所定的範圍內,我們就視它為真實的值而非雜訊。反之在其他情況下,為了能精確地去除雜訊,將做笫二階段的偵測。最後將算出中間值與中間權重式中位數的絕對差值,並將差值和門檻值比較以決定其過濾的結果。
在第二個方法裡,首先我們計算掃描視窗內中間值與其周圍相鄰像素值之間的絕對差異值。並且利用了門檻值的觀念,假如絕對差異值小於門檻值,則將旗標的值設為1,否則將其設為0。接著,我們會加總每個掃瞄視窗內的旗標值,並且稱這個值為”相似度”。我們所設計的濾波器就是利用”相似度”來決定被偵測像素值的過濾結果是用標準中位數、中間權重式中位數濾波器的輸出結果來取代或不改變其值。
我們知道影像的特徵之一就是相鄰的像素值之間都很相似。因此,當局部區域內有脈衝式雜訊出現時,就會顯得很突兀,而所提出的方法就是根據此特性。在第三個方法中,我們首先會在掃瞄視窗內選擇幾個與中間值相鄰的值當作是統計的參考,並利用這些相鄰值的平均值及彼此的平均絕對差異值來計算出一段區間值。這個區間值可以用來表示局部區域內像素值的特性,因此可以將它當作是分辨雜訊點的依據。當被偵測的像素值落在這個區域時,我們就認為它是未受干擾的像素值,所以過濾後的值不變;反之,就是雜訊點並且以中間權重式中位數濾波器的輸出結果來取代其值。
In this thesis, three impulsive noise removal methods for still images are proposed. In the first method, the center pixel value and its surrounding neighbors in the scanning window are sorted. After sorting, depend on the rank of the center pixel value, we decide whether it is a corrupted pixel. If rank falls into the predefined range, center pixel is considered as a normal one. Otherwise, in order to eliminate impulsive noise more accurate, the second detection will be processed. We calculate the absolute difference between center weighted median (weight= 3) and center pixel. The filtering output is according to the result of comparison between the absolute difference and threshold.
In the second method, we calculate the difference between the center pixel and its surroundings neighbors in the scanning window. If the absolute difference is smaller than the predefined threshold, the symbol ‘flag’ is set to 1. Otherwise, it is equal to 0. After that, we sum up the value of flags in the scanning window and call it “similar degree”. The proposed filter is based on the “similar degree” to decide that the center pixel is replaced by the output of standard median, center weighted median or unchanged.
One characteristic of image is that each pixel is very close to surrounding neighbors. If the impulsive noise appears in the local area, it will be very abruptly. The proposed scheme is based on this property. In the third method, we choose several neighboring pixels around the center pixel in the scanning window as the statistic factors. We acquire mean value and average absolute difference among those neighboring pixels and use them to compute a range value. The range value is a representative characteristic of those pixels in the local area and we can use it to identify the noise. If the detected pixel falls into the range, it is considered as an uncorrupted pixel. Otherwise, it is noise and replaced by the output of center weighted median.
Contents
Abstract (in English)........................................I
Abstract (in Chinese).......................................II
AcknowledgementList of Figures.............................III
List of Figures.............................................IV
List of Tables..............................................VI
Contents...................................................VII
Chapter 1 Introduction......................................1
1.1 Motivation..............................................1
1.2 Related Concepts........................................1
1.2.1 The Impulsive Noise Model.........................1
1.2.2 The Mean Filter...................................2
1.2.3 The Standard Median Filter........................4
1.2.4 The Center Weighted Median Filter.................5
1.2.5 The Tri-State Median Filter.......................6
1.2.6 The Noise Adaptive Soft-Switching Median Filter...8
1.2.7 Fast Impulsive Noise Removal......................9
1.3 The Skeleton of Thesis.................................11
Chapter 2 A Two-Stage Switching-Based Median Filter........13
2.1 Introduction...........................................13
2.2 The First Scheme.......................................13
2.3 The Second Scheme......................................20
2.4 Experimental Results...................................22
2.5 Discussions............................................33
Chapter 3 A New Median Filter for Image Denoising UsingLocal Statistics..................................................34
3.1 Introduction...........................................34.
3.2 The Proposed Scheme....................................35
3.3 Experimental Results...................................37
Chapter 4 High Probability Impulsive Noise Removal for Corrupted Images............................................45
4.1 Introductio............................................45
4.2 The Proposed Scheme....................................46
4.3 Experimental Results...................................49
Chapter 5 Conclusions and Future Works.....................60
5.1 Conclusions............................................60
5.2 Future Works...........................................61
References..................................................63
References
中文部分
1.江季龍(2001),適用於影像雜訊消除之濾波器設計,國立中正大學資訊工程研究所碩士論文。
2.謝志彬(2001),數位影像雜訊消除技術之研究,國立中正大學資訊工程研究所碩士論文。
英文部分
3.E. Abreu, M. Lightstone, S. K. Mitra and K. Arakawa (1996), A new efficient approach for the removal of Impulse noise from highly corrupted images, IEEE Transactions on Image Processing, Vol. 5, No. 6, pp. 1012-1025.
4.J. Astola and P. Kuosmanen(1997), Fundamentals of nonlinear digital filtering, Boca Raton, FL: CRC.
5.P. Badulescu and R. Zacin(2000), A two-state switched-median filter, CAS 2000 Proceedings, Vol. 1, pp. 289-292.
6.T. Bose and J. Schroeder (2000), Adaptive mean/median filtering, 2000 IEEE International Conference on Acoustics, Speech, and Signal Processing, Vol. 6, pp. 3830 -3833
7.D.R. K. Brownrigg (1984), The weighted median filter, ACM Communications, Vol. 27, pp. 807-818.
8.T. Chen and H. R. Wu (2000), A new class of median based impulse rejecting filters, Proc. 2000 International Conference on Image Processing, Vol. 1, pp. 916 -919.
9.T. Chen and H. R. Wu (2001), Adaptive impulse detection using center-weighted median filters, IEEE Signal Processing Letters, Vol. 8, No. 1, pp. 1 —3.
10.T. Chen and H. R. Wu (2001), Space variant median filters for the restoration of impulse noise corrupted images, IEEE Transactions on Analog and Digital Signal Processing, Vol. 48, No. 8, pp. 784-789
11.C. T. Chen and L. G. Chen (1996), A self-adjusting weighted median filter for removing impulse noise in image, IEEE International Conference On Image Processing, pp. 419-422.
12.T. Chen, K. K. Ma, and L. H. Chen (1999), Tri-state median filter for image denoising, IEEE Transactions on Image Processing, Vol. 8, No. 12, pp. 1834-1838.
13.D. Florecio and R. W. Schafer (1994), Decision-based median filter using local signal statistics, Proc. SPIE Sysmp. Visual Communications Image Processing, Vol. 2308, pp. 268-275.
14.H.L. Eng and K. K. Ma (2001), Noise adaptive soft-switching median filter, IEEE Transactions on Image Processing, Vol. 10, No. 2, pp. 242-251.
15.H.L. Eng and K. K. Ma (2000), Noise adaptive soft-switching median filter for image denoising, IEEE International Conference on Acoustics, Speech, and Signal Processing, Vol. 4, pp. 2175-2178.
16.S. -J. Ko and Y. H. Lee (1991), Center weighted median filters and their applications to image enhancement, IEEE Transactions on Circuits and Systems, Vol. 38, pp. 984-993.
17.H. Kong and L. Guan (1998), A noise-exclusive adaptive filtering framework for removing impulse noise in digital images, IEEE Transaction on Analog and Digital Signal Processing, Vol. 45, No. 3, pp. 422-428.
18.M. Marl, S. Dellepiane, G. Barragato and C. Venturino (1997), A non-linear adaptive approach for image filtering, Proc. 13th International Conference on Digital Signal Processing, Vol. 1, pp. 303 -306.
19.I. Pitas and A. N. Venetsanopoulos (1990), Nonlinear Digital Filters: Principles and Applications, Boston, MA: Kluwer.
20.G. Pok and J. C. Liu (1999), Decision-based median filter improved by predictions, 1999 International Conference on Image Processing, Vol.2, pp. 410-413.
21.T. Sun and Y. Neuvo (1994), Detail-preserving media based filters in image processing, Pattern Recognition Letter, Vol. 15, pp. 341-347
22.Z. Wang and D. Zhang (1998), Restoration of impulse noise corrupted images using long-range correlation, IEEE Signal Processing Letters, Vol. 5, No. 1, pp. 4 -7
23.Z. Wang and D. Zhang (1999), Progressive switching median filter for the removal of impulse noise from highly corrupted image, IEEE Transactions on circuits and systems II: Analog and Digital Signal Processing, Vol. 46, No. 1, pp 78-80.
24.P. S. Windyga (2000), Fast impulsive noise removal, IEEE Transactions on Image Processing, Vol. 10, No. 1, pp. 173-179.
25.C.B.Wu, B.D. Liu. J. F. Yang and S. C. Hsia (1998), Algorithm of digital impulse noise cancellation, 1998 IEEE Asia-Pacific Conference on Circuits and Systems, pp. 319 -322
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top