Deblurring Images via Dark Channel Prior

Jinshan Pan, Deqing Sun, Hanspeter Pfister, Ming Hsuan Yang

Research output: Contribution to journalArticle

19 Citations (Scopus)

Abstract

We present an effective blind image deblurring algorithm based on the dark channel prior. The motivation of this work is an interesting observation that the dark channel of blurred images is less sparse. While most patches in a clean image contain some dark pixels, this is not the case when they are averaged with neighboring ones by motion blur. This change in sparsity of the dark channel pixels is an inherent property of the motion blur process, which we prove mathematically and validate using image data. Enforcing sparsity of the dark channel thus helps blind deblurring in various scenarios such as natural, face, text, and low-illumination images. However, imposing sparsity of the dark channel introduces a non-convex non-linear optimization problem. In this work, we introduce a linear approximation to address this issue. Extensive experiments demonstrate that the proposed deblurring algorithm achieves the state-of-the-art results on natural images and performs favorably against methods designed for specific scenarios. In addition, we show that the proposed method can be applied to image dehazing.

Original languageEnglish
Article number8048543
Pages (from-to)2315-2328
Number of pages14
JournalIEEE transactions on pattern analysis and machine intelligence
Volume40
Issue number10
DOIs
Publication statusPublished - 2018 Oct 1

Fingerprint

Image Deblurring
Pixels
Sparsity
Motion Blur
Deblurring
Lighting
Pixel
Scenarios
Nonconvex Optimization
Experiments
Linear Approximation
Nonlinear Optimization
Patch
Nonlinear Problem
Illumination
Face
Optimization Problem
Demonstrate
Experiment

All Science Journal Classification (ASJC) codes

  • Software
  • Computer Vision and Pattern Recognition
  • Computational Theory and Mathematics
  • Artificial Intelligence
  • Applied Mathematics

Cite this

Pan, Jinshan ; Sun, Deqing ; Pfister, Hanspeter ; Yang, Ming Hsuan. / Deblurring Images via Dark Channel Prior. In: IEEE transactions on pattern analysis and machine intelligence. 2018 ; Vol. 40, No. 10. pp. 2315-2328.
@article{1b844251b77b4d89884a92dc8bff1819,
title = "Deblurring Images via Dark Channel Prior",
abstract = "We present an effective blind image deblurring algorithm based on the dark channel prior. The motivation of this work is an interesting observation that the dark channel of blurred images is less sparse. While most patches in a clean image contain some dark pixels, this is not the case when they are averaged with neighboring ones by motion blur. This change in sparsity of the dark channel pixels is an inherent property of the motion blur process, which we prove mathematically and validate using image data. Enforcing sparsity of the dark channel thus helps blind deblurring in various scenarios such as natural, face, text, and low-illumination images. However, imposing sparsity of the dark channel introduces a non-convex non-linear optimization problem. In this work, we introduce a linear approximation to address this issue. Extensive experiments demonstrate that the proposed deblurring algorithm achieves the state-of-the-art results on natural images and performs favorably against methods designed for specific scenarios. In addition, we show that the proposed method can be applied to image dehazing.",
author = "Jinshan Pan and Deqing Sun and Hanspeter Pfister and Yang, {Ming Hsuan}",
year = "2018",
month = "10",
day = "1",
doi = "10.1109/TPAMI.2017.2753804",
language = "English",
volume = "40",
pages = "2315--2328",
journal = "IEEE Transactions on Pattern Analysis and Machine Intelligence",
issn = "0162-8828",
publisher = "IEEE Computer Society",
number = "10",

}

Deblurring Images via Dark Channel Prior. / Pan, Jinshan; Sun, Deqing; Pfister, Hanspeter; Yang, Ming Hsuan.

In: IEEE transactions on pattern analysis and machine intelligence, Vol. 40, No. 10, 8048543, 01.10.2018, p. 2315-2328.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Deblurring Images via Dark Channel Prior

AU - Pan, Jinshan

AU - Sun, Deqing

AU - Pfister, Hanspeter

AU - Yang, Ming Hsuan

PY - 2018/10/1

Y1 - 2018/10/1

N2 - We present an effective blind image deblurring algorithm based on the dark channel prior. The motivation of this work is an interesting observation that the dark channel of blurred images is less sparse. While most patches in a clean image contain some dark pixels, this is not the case when they are averaged with neighboring ones by motion blur. This change in sparsity of the dark channel pixels is an inherent property of the motion blur process, which we prove mathematically and validate using image data. Enforcing sparsity of the dark channel thus helps blind deblurring in various scenarios such as natural, face, text, and low-illumination images. However, imposing sparsity of the dark channel introduces a non-convex non-linear optimization problem. In this work, we introduce a linear approximation to address this issue. Extensive experiments demonstrate that the proposed deblurring algorithm achieves the state-of-the-art results on natural images and performs favorably against methods designed for specific scenarios. In addition, we show that the proposed method can be applied to image dehazing.

AB - We present an effective blind image deblurring algorithm based on the dark channel prior. The motivation of this work is an interesting observation that the dark channel of blurred images is less sparse. While most patches in a clean image contain some dark pixels, this is not the case when they are averaged with neighboring ones by motion blur. This change in sparsity of the dark channel pixels is an inherent property of the motion blur process, which we prove mathematically and validate using image data. Enforcing sparsity of the dark channel thus helps blind deblurring in various scenarios such as natural, face, text, and low-illumination images. However, imposing sparsity of the dark channel introduces a non-convex non-linear optimization problem. In this work, we introduce a linear approximation to address this issue. Extensive experiments demonstrate that the proposed deblurring algorithm achieves the state-of-the-art results on natural images and performs favorably against methods designed for specific scenarios. In addition, we show that the proposed method can be applied to image dehazing.

UR - http://www.scopus.com/inward/record.url?scp=85030699983&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85030699983&partnerID=8YFLogxK

U2 - 10.1109/TPAMI.2017.2753804

DO - 10.1109/TPAMI.2017.2753804

M3 - Article

C2 - 28952935

AN - SCOPUS:85030699983

VL - 40

SP - 2315

EP - 2328

JO - IEEE Transactions on Pattern Analysis and Machine Intelligence

JF - IEEE Transactions on Pattern Analysis and Machine Intelligence

SN - 0162-8828

IS - 10

M1 - 8048543

ER -