近年來數位影像處理已有不少應用在安全監控系統中，而清楚
且明暗適中的影像，已經是數位影像的基本需求,但由於攝影環境的
場景複雜，常常會增加影像資訊處理的困難，或者讓處理過後的影
像只有特定部分得到補正而造成功能上的限制，無法符合一般使用
者的需求。
經由模擬的結果顯示出這新式響應曲線與人眼對光的響應曲線
非常相似，都是一個對數型態的轉換函數。這樣的曲線非常適合用
來當作數位影像輸出。除了在影像品質上面做改善，另外在運算處
理速度上也需達到即時的要求，本模擬運算皆由時域轉為頻域運
算，在處理速度上也比時域快很多，因此適用於動態影像之攝影機、
醫療、數位監控系統、車行紀錄和現今的數位相機系統等。
本研究提出一具研究與應用價值的影像強化系統，即將動態範
圍不良的影像校正，對高亮或低亮度之影像作補償及將原本無法辨
識的內容清晰化，以得到正確的影像。

Video enhancement is an important preprocessing phase in many computer vision applications. Many techniques have been proposed to handle various severe environments, In this paper , we proposed a novel method to combine the tone mapping function with the bilateral filter for local contrast enhancement improving ,by our simulation result , this tone mapping curve is similar to human visual as a log transfer function. At same time , the drawback of tone mapping by Gaussian filter is also be improved .

By our experiment result , The proposed algorithm performed well on low contrast images, and the algorithm can be implemented by fast fourier transform for execution efficiency, it will be the suitable function for video camera, driver recorder ,still camera applications .

目 錄
第一章 序論 ............................................ - 1 –
1.1 研究動機  .......................................... - 3 –
1.2 論文架構............................................ - 4 - 

第二章 背景知識與相關研究............................... - 5 - 
2.1影像增強分類(Image Enhancement category)............... - 6 - 
2.2.1 區域性統計增強................................. - 8 - 
2.2.2 雙向濾波器(Bilateral filter)....................... - 10 - 
2.2.3 Homomorphic filter ............... ............... - 13 - 
2.2.4 影像增強相關文獻.............................. - 14 –

第三章 Proposed algorithm................................. - 17 –
3.1 Overview......................................... - 17 –
3.2基本架構......................................... - 18 –
3.3.1 Luminance mapping.............................. - 19 - 
3.3.2 Automatic dodging-and-burning .................... - 22 –
3.3.3 Proposed method...............................– 27 -
第四章 實作後之驗證及結果.............................. - 31 -  
4.1人眼主觀評估.................................... - 31 –
4.2客觀評估........................................ - 33 –
4.3 各個影像處理方式比較............................ - 34 –
4.4 實驗結果........................................ - 36 –

 
第五章 結論與未來研究方向 ............................. - 39 -

參考文獻 .............................................. - 40 –
附錄 .................................................. - 42 –

圖目錄
圖1.1目前在監視系統(Surveillance camera)上所遇到的問題. - 3 –圖2.1影像增強分類...................................... - 7 –
圖2.2 (a)未處理之影像 x(i,j) (b) G(i,j)[x(i,j)-mx(i,j)] (c) 區域性統計增強處理過後之影像 f(i,j) ...................... - 9 –
圖2.3雙向濾波器的輸入與輸出結果........................ - 11 –
圖2.4影像經由雙向濾波器分解的流程...................... - 12 –
圖2.5對稱之高通濾波器.................................. - 13 –
圖2.6 Homomorphic filter 影像增強之流程圖................ - 14 –
圖2.7 (a)處理前之影像，(b)Homomorphic filter 處理後之影像……………………………………………………………………… - 14 –
圖3.1本論文影像增強之基本架構.......................... - 18 –
圖3.2 Grey zone system................................... - 20 –
圖3.3高斯分布(a)以跟(b)差1.6倍並且相減，得到(c) 墨西哥帽濾波器(Mexico hat filter) ...................................... - 23 –
圖3.4經過高斯分布和原始影像做迴旋積之後的Gaussian Pyramid................................................. - 24 –
圖3.5取scale 對影像之影響.............................. - 26 –
圖3.6圖3-6 Halo問題之影像............................. - 26 –
圖3.7對影像發生halo的區域做分析........................ - 27 –
圖3.8邊緣部分的scale均落在scale5...................... - 27 –
圖3.9在scale 5的地方加入bilateral filter.............. - 27 –
圖3.10 (a) center-surround function only(b) center-surround function with Bilateral filter for “halo” improvement......................... - 29 –圖3.11(a) 邊緣發生halo問題  (b) 經過改善後的邊緣........ - 30 –圖4.1(a) Original image  (b) Bilateral filter only  (c) Our proposed method 
.......................... ............................. - 34 –
圖4.2(a) Original image (b) center-surround filter only (c) Our proposed method .................... ............................. - 35 –
圖4.3影像增強前後(應用於監控系統上) .................... - 36 –
圖4.4影像增強前後(應用於數位相機上) .................... - 37 –
圖4.5影像增強前後(應用於行車紀錄器上) ................. - 38 –

表目錄
表3.1 Key值與影亮度的關係.................................. - 20 –
表4.1以投票的方式評估影像之結果...... ..................... - 32 –
表4.2以SVD 對灰階影像評估結果....... ..................... - 33 –

附錄
影像主觀評估.... ........................................... - 42 –

參考文獻

[1.]	T. Arici and Y. Altunbasak, “Image local contrast enhancement using
adaptive non-linear filters,” in IEEE International Conference on Image
Processing (ICIP-06). IEEE, June 2006,

[2.]	“An Efficient Method for Enhancing High-Contrast Digital Photos Automatically”, Jiunn-Lin Wu* and Chia-Yi Chuang, Proceeding of IAENG International Workshop on Imaging Engineering (IWIE'06), Hong Kong, pp. 502 - 506, June 20-22, 2006. (NSC94-2213-E-005-004)

[3.]	Digital Image Processing 3rd Ed. (DIP/3e) by Gonzalez and Woods  © 2008

[4.]	“Gradient Domain High Dynamic Range Compression R. Fattal, D. Lischinski, and M. Werman In ACM Transactions on Graphics, 2002“

[5.]	G. J. Braun and M. D. Fairchild, “Image Lightness Rescaling Using Sigmoidal Contrast Enhancement Function.”, Journal of Electronic Imaging, 8(4), pp. 380, 1999.

[6.]	S. C. Matz, R. J. P. de Figueiredo, “A nonlinear image contrast sharpening approach based on Munsell's scale,” IEEE Transactions on Image Processing, vol.15, no.4, pp.900-909, April 2006.

[7.]	Reinhard, E., Stark, M., Shirley, P., Ferwerda, J.A.: Photographic tone reproduction for digital images. In: Proc. of SIGGRAPH’02,pp. 267–276. ACM Press (2002)

[8.]	Meylan L, Süsstrunk S., High dynamic range image rendering with a Retinex-based adaptive filter. IEEE Trans Image Process. 2006 Sep;15(9):2820-30.

[9.]	LI, Y., SHARAN, L., ADELSON, E. 2005. “Compressing and Companding High Dynamic Range Images with Subband Architectures,” ACM Trans.on Graphics (Proceedings of ACM SIGGRAPH), pp. 836-44

[10.]	 J. Tumblin and G. Turk, “LCIS: A boundary hierarchy for detail-preserving contrast    reduction, ”SIGGRAPH ’99 Annual Conference on Computer Graphics 83-90,Los Angeles,1999

[11.] 基於JND準則實現影像增強. 作者： 謝逸忠.,成功大學工程科學系碩博士班學
位論文/2004年/碩士 (2004/01).

[12.] A. Shnayderman, A. Gusev, and A. M. Eskicioglu, “An SVD-based grayscale image 
quality measure for local and global assessment,” IEEE Trans Image Process. 2006 Feb;15(2):422-9.