臺灣博碩士論文加值系統

近年來數位相機已經變成是生活中的必需品，而且被廣泛地使用來當作影像的擷取裝置。為了達到降低成本與體積的考量，大多數的數位相機都是採用單一圖像感測器搭配一層彩色濾波器矩陣來擷取影像。因為擷取到的影像每一個畫素只有三原色的其中一種顏色，至於其他兩種沒測得的顏色，只好根據相鄰元件所測得的值來推算。此過程就是所謂的解馬賽克(Demosaicking)或是彩色濾波器矩陣內插法(CFA interpolation)。
在本篇論文中提出一個可適性的彩色濾波器矩陣內插法，主要是基於顏色相關性、色差相關性、常色相以及亮度色彩差異性。首先利用一個初始化的步驟大略的估測出遺失的兩種顏色，接著藉由色差相關性和常色相來得到較精確的顏色。由於此種技術一直存有兩種主要的問題，一個是假色問題，另一個是鋸齒現象，這都是因為使用錯誤的演算法而造成的。因此為了得到較佳的影像品質，特別針對這些區域提出了一個可以重複修正的演算法。
經模擬的結果驗證後，我們提出的內插演算法在視覺品質上得到了不錯的影像品質。此外和其他現行的方法比較下，我們提出方法的計算複雜度是可以接受的。

Digital still cameras have become indispensable in people’s life and been widely used as image input devices. In order to reduce cost and size, most commercial digital cameras use a single electronic sensor overlaid with a color filter array (CFA) to capture imagery. Since only one primary color is sampled in each pixel, the missing color primaries must be reconstructed by the method of interpolation from sampled color primaries of the adjacent pixels. This color plane interpolation is known as demosaicking or CFA interpolation. In this paper, an adaptive demosaicking technique using intra-channel correlation, color difference correlation, constant hue, and luminance-color difference correlation is proposed. A rough interpolation based on intra-channel correlation is first implemented by bilinear interpolation. Then the color difference correlation and constant hue are successively used to update the missing color primaries. To obtain high quality color images, color information near the edge regions should be accurately estimated. The correlation between luminance and each color primary is found to be high within a small local region. An adaptive algorithm using luminance-color difference correlation and the information of edge direction is used iteratively for improving the image quality around the edges. Experimental results demonstrate that our proposed algorithm has better image quality in terms of peak signal-to-noise ratio (PSNR) than that obtained by the approach using color difference correlation.

ABSTRACT (in Chinese) Ⅰ
ABSTRACT (in English) Ⅱ
ACKNOWLEDGEMENT Ⅲ
CONTENTS Ⅳ
LIST OF FIGURES Ⅵ
LIST OF TABLES Ⅷ

CHAPTER 1 INTRODUCTION 1
1.1 Background 1
1.1.1 The Structure of a Digital Still Camera 1
1.1.2 Color Filter Array 3
1.1.3 Color Filter Array Interpolation 4
1.2 Motivation 5
1.3 Related work 5
1.4 Objective 7
1.5 Approaches to the Problem 7
1.6 Organization of this Thesis 8
CHAPTER 2 REVIEW OF THE RELATED INTERPOLATION ALGORITHMS 8
2.1 Bilinear Interpolation Method 9
2.2 Edge-sensing Interpolation Method 11
2.3 Smooth-Hue-Transition Interpolation Method 13
CHAPTER 3 THE BASIC IDEA OF THIS THESIS 14
3.1 The Constant Color difference Model 15
3.2 The Constant Hue Model 16
3.3 The Luminance Color Difference Model 18
CHAPTER 4 THE PROPOSED ALGORITHM FOR COLOR FILTER ARRAY
INTERPOLATION 20
4.1 Basic Structure of the Proposed Algorithm 20
4.2 Initial Interpolation 22
4.3 Update of the Green Channel 23
4.4 Update of the Red and Blue Channels 24
4.5 Artifacts that Degrade the Interpolated CFA Images 26
4.6 The Proposed Algorithm for Iterative Update 28
4.6.1 Edge Detection 29
4.6.2 Update R/G/B Channels 30
4.6.3 Stopping Criterion 32
CHAPTER 5 SIMULATION RESULTS 34
5.1 Conditions in Simulation 34
5.2 Fidelity Measure Criterion 35
5.3 Experimental Results 36
CHAPTER 6 CONCLUSIONS 42
REFERENCES 43

[1]B. Bayer, “Color Imaging Array,” U.S. Patent 3 971 065, 1976.
[2]Soo-Chang Pei and Io-Kuong Tam, “Effective color interpolation in CCD color filter array using signal correlation,” IEEE International Conference Image Processing, vol. 3, 2000.
[3]Xin Li, “Demosaicing by successive approximation,” IEEE Transactions Image Processing, vol. 14, no. 3, pp. 370–379, 2005.
[4]Lanlan Chang and Yap-Peng Tan, “Effective use of spatial and spectral correlations for color filter array demosaicking,” IEEE Transactions Consumer Electronics, Vol. 50, pp. 355 - 365, Feb 2004.
[5]R. Kimmel, “Demosaicing: Image reconstruction from CCD samples,”IEEE Transactions Image Processing, vol. 8, pp. 1221–1228, 1999.
[6]D. Cok, “Signal Processing Method and Apparatus for Producing Interpolated Chrominance Values in a Sampled Color Image Signal,”U.S. patent 4 642 678, 1987.
[7]J. Adams, “Interactions between color plane interpolation and other image processing functions in electronic photography,” Proceedings of SPIE, vol. 2416, pp. 144–151, 1995.
[8]B. K. Gunturk, Y. Altunbasak, and R. M. Mersereau, “Color plane interpolation using alternating projections,” IEEE Transactions Image Processing, vol. 11, no. 9, 2002.
[9]Xin Li and M. T. Orchard, “New edge-directed interpolation,” IEEE Transactions Image Processing, vol. 10, no. 10, 2001.
[10]Hung-An Chang; Homer Chen, “Directionally weighted color interpolation for digital cameras,” ISCAS 2005. IEEE International Symposium, vol. 6 pp. 6284 – 6287, 2005.
[11]D. Alleysson, S. Susstrunk, and J. Herault, “Linear demosaicking inspired by human visual system,” IEEE Transactions Image Processing, vol. 14, no. 4, pp. 439–449, 2005.
[12]Naixiang Lian, Lanlan Chang, and Yap-Peng Tan, “Improved color filter array demosaicking by accurate luminance estimation” ICIP 2005. IEEE International Conference, Vol. 1, pp.41-44, 2005.
[13]Xiaolin Wu and Ning Zhang, “Primary-consistent soft-decision color demosaicking for digital cameras,” IEEE Transactions Image Processing, vol. 13, pp. 1263–1274, 2004.
[14]Wenmiao Lu and Yap-Peng Tan, “Color Filter Array Demosaicking:New Method and Performance Measures,” IEEE Transactions Image Processing, Vol. 12, no. 10, october 2003.
[15]T. W. Freeman, “Median Filter for Reconstructing Missing Color Samples,” U.S. Patent 4 724 395, 1988.
[16]J. Hamilton and J. Adams, “Adaptive Color Plane Interpolation in
Single Sensor Color Electronic Camera,” U.S. Patent 5 629 734, 1997.
[17]J. Adams, “Design of practical color filter array interpolation algorithms for digital cameras,” Proceedings of SPIE, vol. 3028, pp. 117–125, Feb. 1997.
[18]http://homepages.inf.ed.ac.uk/rbf/HIPR2/sobel.htm
[19]http://www.r0k.us/graphics/kodak/