Open AccessImage compression system with an optimisation of compression ratio
Author(s) -
Amirjanov Adil,
Dimililer Kamil
Publication year - 2019
Publication title -
iet image processing
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.401
H-Index - 45
eISSN - 1751-9667
pISSN - 1751-9659
DOI - 10.1049/iet-ipr.2019.0114
Subject(s) - image compression , texture compression , compression ratio , data compression , artificial intelligence , data compression ratio , computer science , compression (physics) , image quality , computer vision , color cell compression , artificial neural network , pattern recognition (psychology) , image processing , image (mathematics) , engineering , materials science , composite material , internal combustion engine , automotive engineering
The fundamental goal of image data compression is to set an optimal compression ratio while maintaining an acceptable reproduction quality. This study describes the principles of design of image compression system that automatically sets an optimal compression ratio for particular image content by identifying the image compression method while maintaining a tolerable reproduction quality. The proposed image compression system employs subjective and objective criteria for an assessment of the quality of the image compression system. The supervised artificial neural network is used to identify a compression method among different compression techniques by using subjective criterion. An objective criterion is used to determine an optimal compression ratio, which is calculated by using linear regression analysis that establishes analytical expression between a compression ratio, a property of an image and a reconstructed quality of an image. It was proved that the system is an effective tool for medical image compression applications.