快速影像去模糊演算法

Fast Image Deblurring Algorithms

沈志聰(Chih-Tsung Shen)

指導教授 ： 顏嗣鈞

國立臺灣大學/電機資訊學院/電機工程學研究所/碩士(2008年)

https://doi.org/10.6342/NTU.2008.02529

摘要

現在，許多手持裝置都具備有數位相機模組。大多數的手持裝置小而輕，當拍攝照片時，些許的晃動將很容易的毀壞了這張圖像。許多數位相機公司，如Nikon、Canon、Pentax、Olympus、Sony等，都開發了他們自有的防手振技術。此外，通訊手機公司，如Nokia、Sony-Ericsson、LG，在他們的手持裝置中也都具備有影像穩定方案。

在這篇論文中，我們提出兩個演算法來去除模糊：畫數計量影像去模糊演算法以及加速器輔助影像去模糊演算法。此二演算法，和現存的方法比較起來，有著低耗電及節省時間的優點，還原的影像也具備相當的品質。

其中，畫素計量影像去模糊演算法是軟體解決方案，可以被應用在不同的作業環境以及作業系統上。加速器輔助影像去模糊演算法，紀錄並處理加速器的資料，為一整合硬體及軟體的解決方案。

關鍵字

數位相機 ； 防手振 ； 去模糊 ； 計量 ； 加速器

並列摘要

Nowadays, many handheld devices are equipped with camera modules. As most of the handheld devices are small and light, when taking a picture a slight shake may easily smear the image. Many camera companies, such as Nikon, Canon, Pentax, Olympus, Sony, and more, have developed their own technologies to avoid image blur due to camera shake. Some mobile phone companies, such as Nokia, Sony-Ericsson, and LG, have also incorporated image stabilization solutions in their mobile devices.

In this thesis, two image deblurring algorithms, namely, “Pixel-counting Image Deblurring Algorithm” and “Accelerometer-Aided Image Deblurring Algorithm,” are designed to eliminate motion blur. In terms of power-consumption and time-efficiency, our algorithms are effective in the sense that the produce images of good quality while requiring less power in comparison with existing methods reported in the literature. These two algorithms can operate on PC and easily port on the mobile platform.

The Pixel-Counting Image Deblurring Algorithm is a software solution. It can be used under different platforms and operating systems. The Accelerometer-Aided Image Deblurring Algorithm reads the data file recorded by the accelerometer and then calculates the data. It is an integrated solution of both hardware and software.

並列關鍵字

Digital Camera ； Image Stabilization ； Image Deblurring ； Counting ； Accelerometer

參考文獻

[1] A. C. Bovik, Handbook of Image and Video Processing, Academic Press.

[2] R. Fergus, B. Singh, A. Hertzmann, S. T. Roweis, and W. T. Freeman, “Removing Camera Shake from a Single Image,” in ACM Trans. On Graphics, Volume 25, No. 3, pp.787-794, 2006.

[3] L. Yuan, J. Sun, L. Quan, and H. Y. Shum, ”Image Deblurring with Blurred/Noisy Image Pairs,” in ACM Trans. On Graphics, Volume 26, No. 3, 2007.

[6] A. Agrawal and R. Raskar, “Resolving Objects at Higher Resolution from a Single Motion-blurred Image,” in Proc. of IEEE International Conference on Computer Vision and Pattern Recognition, pp.1-8, 2007.

[8] A. Rav-Acha and S. Peleg, “Two Motion-blurred images are better than one,” in Pattern Recogition Letters, Volume 25, pp.311-317, 2005.