高效率影像視訊超解析度方法

An Efficient Technique for Image/Video Super Resolution

張宸瑋(Chen-Wei Chang)

指導教授 ： 李明穗

國立臺灣大學/電機資訊學院/資訊工程學研究所/碩士(2009年)

https://doi.org/10.6342/NTU.2009.00724

摘要

耗時的處理速度和未考慮影像的內容是現今大部分超解析度方法最嚴重的缺點。因此，我們提出的方法是針對影像中不同的內容，採取不同的方法來處理，藉以大量減少運算時間並同時得到可令人滿意的影像品質。首先，一張單一的影像會先在頻率域中被分割成一個一個八乘八的區塊，然後再對分割出來的區塊加以分類。分類的方法是依據影像中的內容複雜度做為準則，接著再將不同的超解析度方法運用在不同類型的區塊上。我們所提出的方法不僅可以運用在單一影像的解析度提升，更可以被推廣到視訊應用上，不同於單一影像只有空間上的相互關係，視訊還有時間軸的資訊可以被考慮。在一個影像群 (Group Of Picture)中，正在做處理的影像可以參考前面已經被處理過的影像而進一步達到節省更多的計算量。實驗結果與數據顯示，相較於其他的超解析度方法，我們提供出來的方法大幅降低了運算量，同時影像視訊的結果也可以達到令人滿意的效果。

關鍵字

超解析度 ； 區塊

並列摘要

Nowadays, the most serious drawback of super-resolution (SR) is the long processing time and most of methods of SR do not consider the content of the images. Hence, the proposed method focuses on applying different methods to different contents so that the computational complexity can be greatly reduced and the output image quality still remain satisfactory. First, a single image is divided into several 8x8 blocks and then those blocks are classified into three groups in DCT domain, which are flat region, complex area and others. The different methods are, then, processed on different types of blocks. The proposed method is further applied to the video. In a video stream, not only the spatial information but also the temporal correspondence can be taken into consideration. In a group of picture (GOP), the procedure of enhancing the resolution of a block of current frame can utilize the corresponding block of the reference image. The experimental results show that the proposed method provides an efficient way to upsample images and videos with good visual quality.

並列關鍵字

Super-resolution ； Block-based ； Enlargement

參考文獻

[1] H. Stark and P. Oskoui, “High-resolution image recovery from image-plane arrays, using convex projections,” J. Opt. Soc. Am. A, vol. 6, pp. 1715-1726, 1989.

[2] M. Irani and S. Peleg, “Improving resolution by image registration,” Graphical Models and Image Proc., vol. 53, pp. 231-239, May 1991.

[3] R. R. Schultz and R. L. Stevenson, “A Bayesian Approach to Image Expansion for Improved Definition,” IEEE Transactions Image Processing, vol. 3, no. 3, May 1994.

[4] B. K. Gunturk, Y. Altunbasak and R. Mersereau, “Bayesian resolution-enhancement framework for transform-coded video,” Image Processing, 2001. Proceedings. 2001 International Conference, vol. 2, pp. 41-44, Oct. 2001.

[6] B. C. Tom and A. K. Katsaggelos, “Reconstruction of a High Resolution Image from Multiple Degraded Mis-Registered Low Resolution Images,” In SPIE VCIP, vol. 2308, pp. 971–981, Chicago, Sept. 1994.