Image Restoration Using Independent Component Analysis on a Specific Region

• Main Authors: Ting-Chieh Chen, 陳廷杰
• Other Authors: 曾怜玉
• Format: Others
• Language: zh-TW
• Published: 2011
• Online Access: http://ndltd.ncl.edu.tw/handle/71656138518712712440

碩士 === 國立中興大學 === 資訊科學與工程學系所 === 99 === Deblurring is a challenging problem in many digital image processing applications. The sources of degradation can be misfocus, for example many images suffer from lens out of focus blur because of manufacturing limitations etc. A digital image can suffer blurring from a single or a combination of various PSFs(Point Spread Functions). Therefore the blur image restoration becomes a popular issue. Most existing methods require certain knowledge about the prior information, for example the blurring process of image, and their effect depends on the amount of prior information obtained. However, the prior information is usually unknown. Independent component analysis(ICA) is a useful method for recovering signals from their mixtures and can be applied for finding PSF in this paper. Often an iterative procedure is required for estimating the PSF and is computational complex and expensive, also, it is sometimes instable. The method proposed in this thesis is simple and does not require prior information regarding the blurring process. The proposed method has been tested on degraded images and the results are compared to those of existing methods. Experimental results show that the proposed method outperforms some other methods and shows good stability.