Spatially varying blur estimation from a single image. Issue 5 (19th March 2019)

• Record Type: Journal Article
• Authors: El‐Shekheby, Shereen; Abdel‐Kader, Rehab F.; Zaki, Fayez W.
• Is Part Of: IET image processing. Volume 13:Issue 5(2019)
• Page Start: 746
• Page End: 753
• Publication Date: 2019-03-19
• Subjects: edge detection -- image motion analysis -- object detection -- Hough transforms -- image restoration -- image segmentation; detected blur results -- image blur detection results -- blur estimation -- single image -- powerful spatially varying blur detection method -- parametric blur kernel -- input image -- partially blurred images -- blur‐type classification -- structured learning edge‐detection method -- initial blur kernel -- kernel direction -- incorporating -- positive diagonal kernels -- kernel‐specific feature -- initial blur image regions -- reduced edges image segmentation method; Image processing -- Periodicals; 621.36705
• DOI: 10.1049/iet-ipr.2018.5663
• Languages: English
• ISSNs: 1751-9659

Abstract

Abstract : Detection of single image spatially varying blur is a challenging task. Here, the authors propose a powerful spatially varying blur detection method. Blur is detected from a single image without requiring any prior knowledge about the parametric blur kernel, the camera settings, or information about the input image. The proposed approach detects both motion and defocus blurred regions in partially blurred images. First, blur‐type classification is performed using Hough transform and the structured learning edge‐detection method. Second, the initial blur kernel is estimated with a kernel direction that maximises the likelihood of a local window being blurred by incorporating either vertical or positive diagonal kernels. Blur detection is implemented using a kernel‐specific feature. Third, initial blur image regions are refined with the support of the reduced edges image segmentation (CCP) method. Finally, neighbouring information is utilised to refine detected blur results. Comparative results reveal accuracy improvements in the image blur detection results at a reasonable execution time.