Blind deconvolution for single noisy and blurry image using alternating maximum a posteriori estimation with low rank prior

CORC > 沈阳自动化研究所 > 中国科学院沈阳自动化研究所 > 光电信息技术研究室

	Blind deconvolution for single noisy and blurry image using alternating maximum a posteriori estimation with low rank prior
	Sun SJ(孙世杰); Zhao HC(赵怀慈); Lv JF(吕进锋); Hao MG(郝明国); Li B(李波)
	2016
会议名称	2016 16th International Symposium on Communications and Information Technologies (ISCIT)
会议日期	September 26-28, 2016
会议地点	Qingdao, China
关键词	MAP blind deconvolution low rank prior noise
页码	175-181
通讯作者	孙世杰
中文摘要	The purpose of single image blind deconvolution is to estimate the unknown blur kernel from a single observed blurred image and recover the original sharp image. Such task is severely ill-posed and even more challenging especially in the condition that the noise in the input image cannot be negligible. In this paper, the main problem we focus on is how to effectively apply low rank prior to blind deconvolution. A single noisy and blurry image blind deconvolution algorithm is proposed, using alternating maximum a posteriori (MAP) estimation combined with low rank prior. When estimating the intermediate latent image, low rank prior is used as the constraint that is used for noise suppression of the restored image. The denoised intermediate latent image in turn leads to higher quality blur kernel estimation. These two operations are iterated in this manner to arrive at reliable blur kernel estimation. Extensive experiments show the superiority of the proposed method over state-of-the-art techniques, both qualitatively and quantitatively.
收录类别	EI ; CPCI（ISTP）
产权排序	1
会议录	2016 16th International Symposium on Communications and Information Technologies, ISCIT 2016
会议录出版者	IEEE
会议录出版地	New York
语种	英语
ISBN号	978-1-5090-4100-8
WOS记录号	WOS:000391872600037
内容类型	会议论文
源URL	[http://ir.sia.cn/handle/173321/19760]
专题	沈阳自动化研究所_光电信息技术研究室
推荐引用方式 GB/T 7714	Sun SJ,Zhao HC,Lv JF,et al. Blind deconvolution for single noisy and blurry image using alternating maximum a posteriori estimation with low rank prior[C]. 见:2016 16th International Symposium on Communications and Information Technologies (ISCIT). Qingdao, China. September 26-28, 2016.