Method and system for human vision model guided medical image quality assessment

作者: Peter Durlak , Stefan Boehm , Tong Fang , Michelle Xiao-Hong Yan , Markus Lendl

DOI:

关键词: Image qualityEnhanced Data Rates for GSM EvolutionComputer scienceLuminanceQuality (business)Artificial intelligenceRegion of interestContrast (vision)Image compressionComputer visionBlock (programming)

摘要: A method and system for image quality assessment is disclosed. The a no-reference objectively assessing the of medical images. This guided by human vision model in order to accurately reflect perception. region interest (ROI) divided into non-overlapping blocks equal size. Each categorized as smooth block, texture or an edge block. perceptual sharpness measure, which weighted local contrast, calculated each blocks. noise level background luminance, index determined based on measures all blocks, An overall can be using task specific machine learning samples annotated used applications, such video/image compression storage healthcare homeland security, band-width limited wireless communication.

参考文章(12)
Huitao Luo, Assessing image quality ,(2003)
Rony Ferzli, Lina J. Karam, Human Visual System Based No-Reference Objective Image Sharpness Metric international conference on image processing. pp. 2949- 2952 ,(2006) , 10.1109/ICIP.2006.312925
Jacques Dov Barth, System and method for automatic determination of a Region Of Interest within an image Journal of the Acoustical Society of America. ,vol. 121, pp. 694- ,(2003) , 10.1121/1.2640213
I. Tastl, D. Shaked, Sharpness measure: towards automatic image enhancement international conference on image processing. ,vol. 1, pp. 937- 940 ,(2005) , 10.1109/ICIP.2005.1529906
Rolf Holger Wolters, John Hargrove, Wenjing Li, Jia Gu, Method of image quality assessment to produce standardized imaging data ,(2008)