在一般視訊的雜訊量測，是採用擷取一張影像來代表這段影像。可是這並不是個十分合理的現象，因為每一秒就包含了30張的影像。而量測一連串的影像雜訊，更是必需要考慮同一像素的時間關聯性。基於一般影像的雜訊量測使用噪訊比，我們提出的方法是一個有考慮到時間的噪訊比量測方法。

在很多雜訊處理的演算法都有很多取捨，我們更需要一個好的測量的基準。因此這篇論文提出了一些雜訊的量測方法並提供了一種新的測量方式。

When we need to analyze video images, we sometimes capture one still image. But the video is 30 FPS (Frame per Second). But using only one static image to represent this video is unreasonable. Moreover, we must consider the impact of temporal noise. Our method is based on SNR (Signal-to-Noise Ratio) with time relevance of each pixel.

Many noise reduction algorithms have many trade-offs. We need a good measurement method. Therefore this thesis proposes some noise measurement methods and provides one new measurement mode.

口試委員會審定書 i
誌 謝 ii
摘 要 iii
Abstract iv
Content v
Figure Content viii
Table Content ix
Chapter 1 Introduction 1
1.1 Motivation 1
1.2 Noise in Static Image 2
1.2.1 Fixed Pattern Noise 2
1.2.2 Banding Noise 2
1.2.3 Random Noise 3
1.3 Static Image Noise Characteristics 3
1.4 Noise in Video 5
Chapter 2 Previous Noise Measurements 8
2.1 Description 8
2.2 Related Works 8
2.3 Normal SNR (Signal-to-Noise Ratio) 10
2.3.1 Description 10
2.3.2 Formula 10
2.3.3 Pseudo-code 11
2.4 Column Noise 12
2.4.1 Description 12
2.4.2 Formula 12
2.4.3 Pseudo-code 12
2.5 Row Noise 13
2.5.1 Description 13
2.5.2 Formula 13
2.5.3 Pseudo-code 13
2.6 Frame-to-Frame Flicker (FFF) 14
2.6.1 Description 14
2.6.2 Formula 14
2.6.3 Pseudo-code 14
2.7 Temporal Noise 15
2.7.1 Description 15
2.7.2 Formula 15
2.7.3 Pseudo-code 15
2.8 Signal-to-Noise Ratio 16
2.8.1 Description 16
2.8.2 Formula 16
2.8.3 Pseudo-code 16
Chapter 3 Our Method VSNR 17
3.1 Description 17
3.2 Our Method 17
3.2.1 Scene 17
3.2.2 Record 18
3.2.3 Calculate 18
3.3 Formula of VSNR 19
3.4 VSNR Flow Chart 20
3.5 Pseudo code 21
3.6 Down Sample 21
Chapter 4 Experimental Result 23
4.1 Justify 24
4.1.1 Forecast 24
4.1.2 Experimental Result 25
4.2 Vote and Compare with SMIA Spatial SNR. 28
4.2.1 Forecast 28
4.2.2 Experimental Result 29
4.3 Flash Noise Measurement 34
4.3.1 Forecast 34
4.3.2 Experimental Result 34
4.4 Processing Time 39
4.5 Conclusion 40
Chapter 5 Future Work 41
Reference 42

[1] C. Y. Chong, “Image Noise Reduction with Frequency Analysis,” Master Thesis, Department of Computer Science and Information Engineering, National Taiwan University, 2008.
[2] Nokia & ST, “SMIA 1.0,” http://www.kminstrument.co.kr/admin%5Cdata%5Cupload%5CSMIA_Characterisation_Specification.pdf, 2010.
[3] R. M. Haralick and L. G. Shapiro, Computer and Robot Vision, Vol. I, Addison Wesley, Reading, MA, 1992.
[4] Skype, “Skype Hardware Certification Specification,” http://www.megaupload.com/?d=DCRMO1Q6, 2010.
[5] Y. H. Chou, “Video Flash,” http://www.csie.ntu.edu.tw/~fuh/gif/video%20flash.avi, 2010.
[6] Youtube, “Beast,” http://www.youtube.com/watch?v=QSkeB_WfemE, 2010.
[7] Y. Zhao and L. Yu, “A Video Quality Assessment Method,” http://www.megaupload.com/?d=M48G2ZYV, 2010.
[8] Wikipedia, “Signal-to-Noise Ratio,” http://en.wikipedia.org/wiki/Signal-to-noise_ratio, 2010.
[9] Wikipedia, “Wikipedia,” http://en.wikipedia.org/wiki/Wikipedia, 2010.