Improved image defogging method and its application in video monitoring system based on ARM

Purpose – The purpose of this paper is to improve an image defogging algorithm based on a dark channel prior and use this method to clear the foggy image on the Advanced RISC (Reduced Instruction Set Computing) Machines (ARM) platform. Design/methodology/approach – The divided strategy of the foggy...

Full description

Saved in:
Bibliographic Details
Published in:Sensor review 2014-06, Vol.34 (3), p.266-272
Main Authors: Zhu, Haijiang, Yin, Xiupu, Wang, Xuan
Format: Article
Language:English
Subjects:
Algorithms
Automation
Channels
Electrical & electronic engineering
Embedded systems
Engineering
Image enhancement
Image quality
Light
Monitoring
Monitoring systems
Neural networks
Platforms
Strategy
Studies
Video equipment
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Purpose – The purpose of this paper is to improve an image defogging algorithm based on a dark channel prior and use this method to clear the foggy image on the Advanced RISC (Reduced Instruction Set Computing) Machines (ARM) platform. Design/methodology/approach – The divided strategy of the foggy image was proposed through the estimation of the brightness and transmission thresholds. The two regions of the foggy image were processed using two transmissions. Many foggy images were tested using the improved method. Findings – The finding resulting from this study is that a divided strategy has been proposed to use the image defogging. Compared with the existing methods, the running time of the improved method is less on the ARM platform. Practical implications – Image enhancement is an important technology of digital images, and the quality of images plays a key role in the video monitoring and the intelligent transportation system. Originality/value – This paper presented an improved image defogging method using the divided strategy and a substantial number of experimental results was provided to demonstrate this method.
ISSN:0260-2288
1758-6828
DOI:10.1108/SR-05-2012-684