跳到主要內容

臺灣博碩士論文加值系統

(3.229.124.74) 您好!臺灣時間:2022/08/11 07:07
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:黃仲裕
研究生(外文):Chung-Yu Huang
論文名稱:不同光線下彩色影像色彩校準之研究
論文名稱(外文):The Study of Color Corrections for Color Images under Varying Illuminants.
指導教授:江政欽江政欽引用關係
指導教授(外文):Cheng-Chin Chiang
學位類別:碩士
校院名稱:國立東華大學
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:2002
畢業學年度:90
語文別:中文
論文頁數:78
中文關鍵詞:電腦視覺光源顏色恆常性切割主成分分析校準
外文關鍵詞:illuminantcomputer visioncolor constancysegmentprincipal component analysiscorrect
相關次數:
  • 被引用被引用:8
  • 點閱點閱:792
  • 評分評分:
  • 下載下載:251
  • 收藏至我的研究室書目清單書目收藏:2
在電腦視覺中,顏色是一個很重要的特徵,但是影像中的顏色總是受到光源的影響,而色彩恆常性就是用來解決這類的問題。我們提出一個與色彩恆常性不同的方法來解決這類問題。我們的做法與色彩恆常性一個明顯不同之處在於我們將影像依顏色不同做區塊切割,而且也導致更好的效果。另外一個不同之處是在校準實驗中我們會利用一張在正常光源下所拍攝的影像當作較準的參考影像(我們稱之為標準影像),另外我們稱在不同光源下所拍攝的待校準影像影像為測試影像。接著我們藉由標準影像中的顏色使用主成分分析來校準測試影像中的顏色。當測試影像中有標準影像沒有出現的前景,我們也可以用背景來校準它的顏色。實驗結果顯示這種做法確實具有很好的校準效果,可應用於一些彩色影像擷取或輸出裝置的色彩校準。
In computer vision applications, color is a very important feature. However, the colors in the images are very sensitive to the illuminants. Some color constancy methods are proposed to handle this kind of sensitivities. In this thesis, we propose a different method to solve the problem of color corrections under varying illuminants. One of the most different features between the previous color constancy methods and ours is that we segment the testing image into regions by color. This way leads to satisfactory performance. The other difference is that we use a standard image as the reference image for color correction. The images captured under different illuminants are called testing images. The principal component analyses are performed both on standard image and testing image. Then, the transformation between these two principal components is then calculated. The calculated transformation is what we need to correct the colors in testing image. With this method, even for the foreground objects not appearing in the standard image, we still can correct its colors from the color of some background in the standard image. According to our experimental results, the performance is very satisfactory. This method can be applied to the color corrections of some color image acquisition and output devices.
摘要 2
ABSTRACT 3
致謝(Acknowledgement) 4
目錄 6
圖目錄 7
表目錄 8
第 1 章 導論 9
1.1 研究動機與目的 9
1.2 研究背景 10
1.2.1 色彩學 10
1.3 色彩恆常性的方法 14
1.3.1 灰階界(Gray World) 15
1.3.2 色域圖對映(Gamut Mapping) 16
1.3.3 顏色透視(Color in Perspective) 17
1.3.4 類神經網路(Neural Networks) 18
1.3.5 由關係性決定顏色(Color by Correlation) 19
第 2 章 實驗方法 20
2.1 影像資料庫 20
2.2 採用的色彩空間(Color Space) 22
2.2.1 B-Y/R-Y空間 22
2.2.2 調適性的rg空間 23
2.2.3 RGB空間 26
2.3 使用主成分分析(Principal Component Analysis)校準 27
2.3.1 平移(Translation) 28
2.3.2 旋轉(Rotation) 29
2.3.3 縮放(Scaling) 29
2.3.4 再旋轉 29
2.3.5 再平移 30
2.4 以影像區域為主(Region Based)之色彩校準 31
2.5 亮度的加強 33
第 3 章 實驗結果分析 35
3.1 分析是否對實驗結果的亮度加強與各色彩空間的比較 36
3.2 分析是否要對實驗中無前景的影像做區域的切割 41
3.3 分析是否要對實驗中有前景的影像做區域的切割 47
3.4 不同色溫下標準色光的各色彩空間比較 56
3.5 實際應用:人臉在不同顏色的光源下的校正 61
第 4 章 結論與未來研究方向 63
4.1 結論 63
4.2 未來研究方向 65
參考文獻 66
附錄 69
[1]中井義雄 and 川崎秀昭, “現代色彩學 Modern Color Co-ordination Theory,” 日本色研事業株式會社, 1999年9月。
[2]http://ww2.philips.com.tw/lighting/p-fram.html
[3]Kobus Barnard, “Modeling Scene Illumination Colour for Computer Vision and Image Reproduction: A survey of computational approaches,” SFU Ph.D. depth paper , December 1998.
[4]G. D. Finlayson, S. D. Hordley, and P. M. Hubel, “Colour by Correlation: A Simple, Unifying Approach to Colour Constancy,” Computer Vision, 1999. The Proceedings of the 7th IEEE International Conference, Volume: 2, Pages: 835-842, 1999.
[5]http://cie.kee.hu/newcie/publ/abst/s005.html
[6]Kobus Barnard, “Improvements to Gamut Mapping Colour Constancy Algorithms,” ECCV’2000 Proceedings 6th European Conference on Computer Vision, Dublin, Pages: 390-402, 2000.
[7]Donald Hearn and M. Pauline Baker, “Computer Graphics C Version”, Second Edition, Prentice Hall, Chapter 14, Pages: 505-506, 1997.
[8]G. D. Finlayson, “Color in Perspective,” IEEE Transaction on Pattern Analysis and Machine Intelligence, Volume: 18, Issue: 10, Pages: 1034-1038, October, 1996.
[9]http://home.kimo.com.tw/lzraqua/lamp-price.htm
[10]http://www.nelt.co.jp/navi/index.html
[11]http://www.toa.com.tw/Web/chinese/light_04.php
[12]Soo-Chang Pei and Ching-Long Tseng, “Face Detection for Difference Chromatic Illuminations,” 14th Conference on Computer Vision, Graphics and Image Processing, Kanding, Pingtung, Taiwan, Aug., 2001.
[13]K. R. Rao and J. J. Hwang, “Techniques and Standards for Image, Video, and Audio Coding,” Prentice Hall, Chapter 2, Pages: 9-16, 1997.
[14]Brian Funt, Kobus Barnard and Lindsay Martin, “Is Machine Colour Constancy Good Enough?” 5th European Conference on Computer Vision, Pages: 445-459, 1998.
[15]M. Soriano, E. Marszalec, M. Pietikäinen, “Color Correction of Face Images Under Different Illuminants by RGB Eigenfaces,” Proc. 2nd Audio- and Video-Based Biometric Person Authentication Conference (AVBPA99), March 22-23, Washington DC, USA, Pages: 148-153, 1999.
[16]Simon Haykin, “Neural Networks: A Comprehensive Foundation,” Second Edition, Prentice Hall, 1999.
[17]Günther Wyszecki and W. S. Stiles, “Color Science: Concepts and Methods, Quantitative Data and Formulae,” Second Edition, Wiley Inter-Science, Chapter 3, Pages: 139, 164-166, August, 2000.
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top