Linear Gaussian blur evolution for detection of blurry images

E. Tsomko, Hyong Joong Kim, E. Izquierdo

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

Even though state-of-the-art digital cameras are equipped with auto-focusing and motion compensation functions, several other factors including limited contrast, inappropriate exposure time and improper device handling can still lead to unsatisfactory image quality such as blurriness. Indeed, blurry images make up a significant percentage of anyone's picture collections. Consequently, an efficient tool to detect blurry images and label or separate them for automatic deletion in order to preserve storage capacity and the quality of image collections is needed. A new technique for automatic detection and removal of blurry pictures is presented. Initially, a set of interest points and local image areas is extracted. These areas are then evolved in time according to the conventional linear scale space. The gradient of the evolution curve through scale is then used to produce a 'blur graph' representing the probability of a picture being blurred or not. Complexity is kept low by applying a Monte-Carlo like technique for the selection of representative image areas and interest points and by implicitly estimating the gradient of the scale-space curve evolution. An exhaustive evaluation of the proposed technique is conducted to validate its performance in terms of detection accuracy and efficiency.

Original languageEnglish
Article numberIIPEAT000004000004000302000001
Pages (from-to)302-312
Number of pages11
JournalIET Image Processing
Volume4
Issue number4
DOIs
Publication statusPublished - 2010 Aug 1

Fingerprint

Motion compensation
Digital cameras
Image quality
Labels

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition
  • Signal Processing
  • Software
  • Electrical and Electronic Engineering

Cite this

Linear Gaussian blur evolution for detection of blurry images. / Tsomko, E.; Kim, Hyong Joong; Izquierdo, E.

In: IET Image Processing, Vol. 4, No. 4, IIPEAT000004000004000302000001, 01.08.2010, p. 302-312.

Research output: Contribution to journalArticle

Tsomko, E, Kim, HJ & Izquierdo, E 2010, 'Linear Gaussian blur evolution for detection of blurry images', IET Image Processing, vol. 4, no. 4, IIPEAT000004000004000302000001, pp. 302-312. https://doi.org/10.1049/iet-ipr.2009.0001
Tsomko, E. ; Kim, Hyong Joong ; Izquierdo, E. / Linear Gaussian blur evolution for detection of blurry images. In: IET Image Processing. 2010 ; Vol. 4, No. 4. pp. 302-312.
@article{979bd4d0daa242dd8915aab3722075d8,
title = "Linear Gaussian blur evolution for detection of blurry images",
abstract = "Even though state-of-the-art digital cameras are equipped with auto-focusing and motion compensation functions, several other factors including limited contrast, inappropriate exposure time and improper device handling can still lead to unsatisfactory image quality such as blurriness. Indeed, blurry images make up a significant percentage of anyone's picture collections. Consequently, an efficient tool to detect blurry images and label or separate them for automatic deletion in order to preserve storage capacity and the quality of image collections is needed. A new technique for automatic detection and removal of blurry pictures is presented. Initially, a set of interest points and local image areas is extracted. These areas are then evolved in time according to the conventional linear scale space. The gradient of the evolution curve through scale is then used to produce a 'blur graph' representing the probability of a picture being blurred or not. Complexity is kept low by applying a Monte-Carlo like technique for the selection of representative image areas and interest points and by implicitly estimating the gradient of the scale-space curve evolution. An exhaustive evaluation of the proposed technique is conducted to validate its performance in terms of detection accuracy and efficiency.",
author = "E. Tsomko and Kim, {Hyong Joong} and E. Izquierdo",
year = "2010",
month = "8",
day = "1",
doi = "10.1049/iet-ipr.2009.0001",
language = "English",
volume = "4",
pages = "302--312",
journal = "IET Image Processing",
issn = "1751-9659",
publisher = "Institution of Engineering and Technology",
number = "4",

}

TY - JOUR

T1 - Linear Gaussian blur evolution for detection of blurry images

AU - Tsomko, E.

AU - Kim, Hyong Joong

AU - Izquierdo, E.

PY - 2010/8/1

Y1 - 2010/8/1

N2 - Even though state-of-the-art digital cameras are equipped with auto-focusing and motion compensation functions, several other factors including limited contrast, inappropriate exposure time and improper device handling can still lead to unsatisfactory image quality such as blurriness. Indeed, blurry images make up a significant percentage of anyone's picture collections. Consequently, an efficient tool to detect blurry images and label or separate them for automatic deletion in order to preserve storage capacity and the quality of image collections is needed. A new technique for automatic detection and removal of blurry pictures is presented. Initially, a set of interest points and local image areas is extracted. These areas are then evolved in time according to the conventional linear scale space. The gradient of the evolution curve through scale is then used to produce a 'blur graph' representing the probability of a picture being blurred or not. Complexity is kept low by applying a Monte-Carlo like technique for the selection of representative image areas and interest points and by implicitly estimating the gradient of the scale-space curve evolution. An exhaustive evaluation of the proposed technique is conducted to validate its performance in terms of detection accuracy and efficiency.

AB - Even though state-of-the-art digital cameras are equipped with auto-focusing and motion compensation functions, several other factors including limited contrast, inappropriate exposure time and improper device handling can still lead to unsatisfactory image quality such as blurriness. Indeed, blurry images make up a significant percentage of anyone's picture collections. Consequently, an efficient tool to detect blurry images and label or separate them for automatic deletion in order to preserve storage capacity and the quality of image collections is needed. A new technique for automatic detection and removal of blurry pictures is presented. Initially, a set of interest points and local image areas is extracted. These areas are then evolved in time according to the conventional linear scale space. The gradient of the evolution curve through scale is then used to produce a 'blur graph' representing the probability of a picture being blurred or not. Complexity is kept low by applying a Monte-Carlo like technique for the selection of representative image areas and interest points and by implicitly estimating the gradient of the scale-space curve evolution. An exhaustive evaluation of the proposed technique is conducted to validate its performance in terms of detection accuracy and efficiency.

UR - http://www.scopus.com/inward/record.url?scp=77955532668&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77955532668&partnerID=8YFLogxK

U2 - 10.1049/iet-ipr.2009.0001

DO - 10.1049/iet-ipr.2009.0001

M3 - Article

AN - SCOPUS:77955532668

VL - 4

SP - 302

EP - 312

JO - IET Image Processing

JF - IET Image Processing

SN - 1751-9659

IS - 4

M1 - IIPEAT000004000004000302000001

ER -