壓縮影像還原、圖片降噪暨畫質提升的自適應濾波技術

在這個資訊爆炸的時代，隨著多媒體以及網路通訊技術的發展，廠商和消費者不斷追求更高畫質以及更高的壓縮率，甚至於更多更廣泛的多媒體運用，資訊流量到達了一個前所未有的高峰。然而有一些問題卻是一直存在的，我們不想見到的雜訊以及壓縮技術過程產生的壓縮失真及不自然痕跡。例如在圖像壓縮的時候，為了節省空間，會使用高壓縮比的失真壓縮來處理，而這種失真壓縮會產生許多影響圖片品質的不自然痕跡，類似的不自然痕跡跟失真在影片壓縮領域也很常見。另一方面，在數位相片拍攝的過程中，會有因為環境因以及感光元件特性而造成的雜訊，影響了成像的品質，或是在影像訊號類比轉換及傳輸的過程，難免會有訊息損失或是雜訊產生，例如常見的類比電視的訊號便是一例。　　本論文提出了一個相當有效的演算法，以降低不自然痕跡以及雜訊，這個演算法改編自雙邊濾波器，並且結合了許多不同畫質提升除噪演算法的特性，具有相當優良的自適應性，並且提出了針對彩色影像的色域轉換處理方式，以及改良了超畫素分割法，能夠適應不同材質的影像，效果遠勝於目前知名的方法，達到快速且極為優良的畫質提升以及降低不自然痕跡與雜訊的效果。

關鍵字

自適應濾波器；降噪；不自然痕跡消除

並列摘要

In recent years, the rapid growth of communication technology and internet has pushed the multimedia to a brand new era. The more complex the multimedia is, the more demanding the requirement is. People nowadays are in pursuit of higher quality of multimedia. The bandwidth of information has reached its climax ever; however, there are still some certain existing problems. For example, we absolutely would not like to spot the noise, artifacts and image distortion, some of which especially occurred in the compressing process. Because when compressing the image, we use high compression ratio to reduce file size, or to save transmission bandwidth, leading to artifacts that could have great influence on the quality of the images. On the other hand, in the process of acquiring digital images, the noises caused by environmental factors and image sensors may affect the quality of images. Also, during the conversion, decoding, transmission and receiving process of analogue signals, they will unavoidably bring about the loss of signals and the production of noises, like what we can easily observe on analogue television. 　　In this thesis, we propose a useful and powerful algorithm to reduce artifacts and noise with low complexity. Firstly, the algorithm is adapted from bilateral filter and is combined with the characteristic of some famous denoising algorithms. Secondly, we proposed a disclosure of processing method on the transformation of color space for color images. Last but not the least, when the adapted SLIC superpixel segmentation are applied to various textures of image, the result is far effective from renowned method up to present. Therefore, it can enhance the image quality while reduce artifacts produced by compression process and eliminate Gaussian noise from images effectively with low complexity.

並列關鍵字

adaptive filterl ； noise reduction ； artifact removal

參考文獻

[1] G. Wallace, “The JPEG still picture compression standard,” Commun. ACM, vol. 34, no. 4, pp. 30–44, 1991.

[2] D. Taubman, M. Marcellin, “JPEG2000: Image Compression Fundamentals, Standards, and Practice,” Norwell, MA: Kluwer, 2001.

[3] Huynh-Thu, Q., Ghanbari, M., "Scope of validity of PSNR in image/video quality assessment," Electronics Letters , vol.44, no.13, pp.800,801, June 19 2008

[4] Zhou Wang, Alan C. Bovik, Hamid R. Sheikh, and Eero P. Simoncelli, "Image quality assessment: from error visibility to structural similairty," IEEE Transactions on Image Processing, vol. 13, no. 4, pp. 600−612, Apr. 2004.

[9] T. Huang, G. Yang, and G. Tang, "A fast two-dimensional median filtering algorithm," IEEE Trans. Acoust., Speech, Signal Processing, vol. 27, no. 1, pp. 13–18, 1979.

國際替代計量

壓縮影像還原、圖片降噪暨畫質提升的自適應濾波技術

全文下載

主題瀏覽