Home // EMERGING 2017, The Ninth International Conference on Emerging Networks and Systems Intelligence // View article

Two-step Facial Images Deblurring With Kernel Refinement Via Smooth Priors

Authors:
Jun-Hong Chen
Long-Wen Chang

Keywords: image debluring; kernel; smooth prior

Abstract:
Image deblurring is a challenging task in image processing. It is an ill-posed problem to estimate the unknown blur kernel and recover the original image from a blurred image. There are many methods for blurred natural images; however, few of them are able to perform well on blurred face images. Based on L_0 norm prior, we propose a two-step method for the images deblurring. The proposed method does not require any facial dataset to initialize the gradient of contours or any complex filtering strategies. In the first step, we combine L_0 norm prior with our local smooth prior to predict the blur kernel. With simple Gaussian filtering, we could maintain the smooth region in the latent image. In the second step, we refine the previous estimated kernel. In order to discard low intensity pixels that seemed to be noises on the kernel, we impose the sparsity on the kernel. Experimental results demonstrate that our proposed algorithm performs well on the facial images.

Pages: 21 to 24

Copyright: Copyright (c) IARIA, 2017

Publication date: November 12, 2017

Published in: conference

ISSN: 2326-9383

ISBN: 978-1-61208-602-6

Location: Barcelona, Spain

Dates: from November 12, 2017 to November 16, 2017