臺灣博碩士論文加值系統

&;#63849;位影像常常由於大氣擾動等一些外在因素或是在擷取與傳輸影像的過程中產生雜訊，而為了因應這些雜訊問題，許多處理雜訊的方法也陸續被提出，在此本篇研究則是針對高斯雜訊進行處理。由於傳統處&;#63972;方式，如平均&;#63876;波器，雖然方法簡單，但是容&;#63968;造成影像邊緣與細節過&;#64001;的模糊。因此後來發展的雜訊處理方式，改進的方向在於去除雜訊的同時還能有效的保&;#63949;影像細節與邊緣，然而有些方法雖然有效但是需要花費較長的處理時間。而如何準確的判斷出是雜訊或是影像細節則是關係到影像品質的重大關鍵。因此後來有學者提出高斯雜訊標準差的估計方式，透過參數的設定調整來進行研究。並使用自動化的方式取得估計的高斯雜訊標準差，如此一來便能夠有效的利用估計的高斯雜訊標準差進行影像處理，但是此方法在估計雜訊時需要花費一些時間。因此本篇研究提出一個簡易型的雙向濾波器(Simplified Bilateral Filter)，相對於傳統的雙向濾波器能夠花費較少的時間有效的去除雜訊並保留影像邊緣。另外我們透過建立Lookup-Table減少計算權重的指數運算時間。也提出對數搜尋法(Logarithm Search Method)的概念來加速雜訊估計的時間。從實驗結果來看，本方法不但能夠準確的估計高斯雜訊標準差，在影像處理的結果看來也能有效的去除雜訊並保留影像邊緣與細節，更重要的是在整體的速度上能夠達到即時的應用，因此非常適合應用於硬體相關實作。

Digital images are often corrupted by noise during image acquisition or transmission due to a number of nonidealities encountered in image sensors and communication channels. To overcome this problem, many methods were proposed. In this study, we focus on Gaussian noise. Traditional methods such that mean filter is simple, but it often causes blurred edge. Therefore many methods were proposed for noise reduction and preserving image edge. However, most of these methods need more time for processing noisy image. Besides it has one major problem that it is not easy to decide a pixel is an image detail or a noise. To overcome this problem, someone proposed automatic estimation method for Gaussian noise standard deviation by parameter tuning. And we can use this estimation value for image processing. But these are some problems that it needs lot of time for Gaussian noise standard deviation estimation. To solve this problem we proposed a fast algorithm called Logarithm Search Method (LSM). In this study we propose a simplified bilateral filter (SBF). SBF uses less time to remove noise and preserve image edge. Then we use Lookup-Table to reduce the time for calculate weight. In experiment result, we can observe that our proposed method efficient remove noise while preserving image edge. The most important is that it can filter the noisy image in real-time. As a result, the proposed method is suitable for hardware implementation.

使用簡易型雙向濾波器消除影像高斯雜訊 i
Gaussian Noise Removal Using Simplified Bilateral Filtering ii
目錄 iii
圖目錄 iv
表目錄 vi
第一章 諸論 1
1.1 研究動機與背景 1
1.2 論文架構 2
第二章 文獻回顧 3
2.1 平均濾波器(Mean Filter) 4
2.2 雙向濾波器(Bilateral Filter) 6
2.3 Non-Local Means (NL-means) 7
2.4 線性分段模式(Piecewise linear models) 9
2.5 Noise Samples Based Method (NSBM) 13
2.6 Edge Map Based Method (EMBM) 15
2.7影像品質評估 18
第三章 使用簡易型雙向濾波器消除高斯雜訊 19
3.1 Logarithm Search Method (LSM) 19
3.2 簡易型雙向濾波器(Simplified Bilateral Filter(SBF)) 24
3.3 Lookup Table 27
第四章 實驗結果與討論 28
4.1 人工測試影像的實驗結果 28
4.2 真實影像的實驗結果 39
第五章 結論與未來工作 46
參考文獻 47

[1]. I. Pitas and A. N. Venetsanopoulos, “Nonlinear mean filters in image processing,” IEEE Trans. Acoust., Speech, Signal Processing vol. ASSP-34, pp. 573-584, June 1986
[2]K. Arakawa, “Median filter based on fuzzy rules and its application to image restoration,” Fuzzy Sets and Systems, vol. 77, pp. 3-13, 1996.
[3]R. Lukac, “Adaptive vector median filtering,” Pattern Recogn. Lett, vol. 24, pp. 1889-1899, 2003.
[4]J. Y. Chang and J. L. Chen, “Classified-augmented median filters for image restoration,” IEEE TINSTRUM MEAS, vol. 53, no. 2, pp. 351-356, 2004.
[5]M. I. Vardavoulia, I. Andreadis, and P. Tsalides, “A new vector median filter for color image processing,” Pattern Recogn. Lett, vol. 22, no. 6-7, pp. 675-689, 2001.
[6]Pei-Eng Ng and Kai-Kuang Ma, ”A Switching median filter with boundary descriminative noise detection for extremely corrupted images,” IEEE Trans. Image Processing, vol.15,2006
[7]E. J. Balster, Y. F. Zheng, and R. L. Ewing, “Feature-based wavelet shrinkage algorithm for image denoising,” IEEE Trans. Image Process., vol. 14, no. 12, pp. 2024–2039, Dec. 2005.
[8]B. Zhang, J. M. Fadili, and J. L. Starck, “Wavelets, ridgelets, and curvelets for poisson noise removal,” IEEE Trans. Image Process., vol. 17, no. 7, pp. 1093–1108, Jul. 2008.
[9]C. Tomasi and R. Manduchi, “Bilateral filter for gray and color images,” in Proc. IEEE Int. Conf. Computer Vision, 1998, pp. 839–846.
[10] A. Buades, B. Coll, and J.M Morel, “A review of image denoising algorithms, with a new one,” SIAM-Multiscale Modeling and Simulation, 4:490–530, 2005.
[11]Jiunn-Lin Wu* and Ho-Yu Chen, “High ISO Noise Reduction Using Simplified Bilateral Filtering,” 2009 Workshop on Consumer Electronics, Taichung, November 6 , 2009.
[12]F. Russo, “A method for estimation and filtering of Gaussian noise in images,” IEEE Trans. Instrum. Meas., vol. 52, no. 4, pp. 1148–1154, Aug. 2003.
[13] F. Russo,“Automatic enhancement of noisy images using objective evaluation of image quality,” in Proc. IEEE IMTC, Como, Italy, May 18–20, 2004, pp. 621–626.
[14] F. Russo, “Image filtering based on piecewise linear models,” in Proc. IEEE Int. Workshop IST, Stresa, Italy, May 2004, pp. 7–12.
[15]F. Russo, “A method based on piecewise linear models for accurate restoration of images corrupted by Gaussian noise,” IEEE Trans. Instrum. Meas., vol. 55, no. 6, December 2006, pp.1935-1943.
[16] Angelo Bosco. Arcangelo Bruna. Giuseppe Messina. And Giuseppe Spampinato. “Fast Method for Noise Level Estimation and Integrated Noise Reduction,” IEEE Trans. Consum. Electron., vol. 51, No. 3, AUGUST 2005, pp.1028-1033.