在這篇論文，我們介紹一個兩階段的影像去雜訊演算法，包含主成分分析的局部像素分組(LPG_PCA)和相鄰嵌入(NE)兩種方法。在LPG_PCA方法中，我們從一張有雜訊的影像將局部相似的補丁分組，然後利用這些補丁透過PCA技術分析顯著的成分。當PCA域中的係數收縮後，大部分的雜訊可以被抑制。這個階段所產生的雜訊殘留影像及被估計的雜訊程度被用於第二階段。在NE演算法中，我們需要一至兩張無雜訊的影像，及其對應的雜訊影像(要求雜訊的程度與輸入的雜訊影像相同)當作訓練集。透過將影像切割成固定大小的補丁並比對後，我們可以參考雜訊影像補丁間的關係，重建出去雜訊影像的補丁，並產生去雜訊的影像。
最後，我們的實驗結果證明：我們提出的方法不僅可以改善原本的去雜訊演算法，而且在視覺品質和客觀標準上都獲得更好的去雜訊表現。

This paper introduces a two-stage image denoising method which consists principal component analysis with local pixel grouping (LPG_PCA) and neighbor embedding (NE). We group local similar patches from input noise image, and then we use PCA technique to analysis the significant component by these patches. After shrinking the coefficient in the PCA domain, most of noise can be suppressed. The denoised image by LPG_PCA and estimated noise level are taken for second stage. In NE algorithm, we find several nearest patches for each input noisy image patch from the noisy training images and compute the reconstruction weights, and then we use the weights to reconstruct the denoised image patches through the support of the noise-free training images.
Finally, the experimental results demonstrate our method not only improves the original algorithms but also achieves better denoising performance both the visual quality and the objective criteria such as PSNR and SSIM.

[1] A. Buades, B. Coll, and J-M. Morel, “A non-local algorithm for image denoising,” in Proc. IEEE International Conference. Computer Vision Pattern Recognition, vol. 2, pp. 60–65, Jun. 2005.
[2] A. Buades, B. Coll, and J-M. Morel, “Nonlocal image and movie denoising,” International Journal of Computer Vision, vol. 76, no. 2, pp. 123–139, Feb. 2008.
[3] A. Buades, B. Coll, and J-M. Morel, “Non-local means denoising,” Image Processing On Line, vo1. 2011.
[4] Y. Chen and K. J. R. Liu, “ A game theoretical approach for image denoising,” in IEEE International Conference on Image Processing, pp. 1125–1128, Sep. 2010.
[5] P-C. Hsiao, L-W. Chang, “Image denoising with dominant sets by a coalitional
game approach,” IEEE Transactions on Image Processing, vol. 22, no. 2, pp. 724–738, Feb. 2013.
[6] M. Elad, M. Aharon, “Image denoising via sparse and redundant representations over learned dictionaries,” IEEE Transactions on Image Processing, vol. 15, no. 12, pp. 3736–3745, Dec. 2006.
[7] P. Chatterjee and P. Milanfar, “Clustering-based denoising with locally learned dictionaries,” IEEE Transactions on Image Processing, vol. 18, no. 7, pp. 1438–1451, July 2009.
[8] H. Takeda, S. Farsiu, and P. Milanfar, "Kernel regression for image processing and reconstruction," IEEE Transactions on Image Processing, vol. 16, no. 2, pp. 349–366, Feb. 2007.
[9] K. Dabov, A. Foi, V. Katkovnik, and K. Egiazarian, “Image denoising by sparse 3-D transform-domain collaborative filtering,” IEEE Transaction on Image Processing, vol. 16, no. 8, pp. 2080–2095, Aug. 2007.
[10] D.D. Muresan and T.W. Parks, “Adaptive principal components and image denoising,” in IEEE International Conference on Image Processing, vol. 1, pp. 101–104, Sept. 2003.
[11] K. Dabov, A. Foi, V. Katkovnik, and K. Egiazarian, “BM3D image denoising with shape-adaptive principal component analysis,” in SPARS-Signal Processing with Adaptive Sparse Structured Representations, Saint Malo, France, Apr. 2009.
[12] L. Zhang, W. Dong, D. Zhang, and G. Shi, “Two-stage image denoising by principal component analysis with local pixel grouping,” Pattern Recognition, vol. 43, no. 4 pp. 1531–1549, Apr. 2010.
[13] R. Shi, I-F Shen, and W. Chen, “Image denoising through locally linear embedding,” in Proc. CGIV-Computer Graphics, Imaging and Vision: New Trends, pp. 147-152, July 2005.
[14] Z. Wang, A. C. Bovik, H. R. Sheikh, and E. P. Simoncelli, “Image quality assessment: From error visibility to structural similarity,” IEEE Transaction on Image Processing, vol. 13, no. 4, pp. 600–612, Apr. 2004.