سال انتشار: ۱۳۹۰

محل انتشار: هفتمین کنفرانس ماشین بینایی و پردازش تصویر

تعداد صفحات: ۴

نویسنده(ها):

Abdolrahman Attar – Department of Computer EngineeringFaculty of Engineering University of Guilan, Rasht, Iran P.O.Box: 3756-41635,
Reza Moradi Rad – Department of Computer EngineeringFaculty of Engineering P.O.Box: 3756-41635,University of Guilan, Rasht, Iran
Asadollah Shahbahrami – Department of Computer EngineeringFaculty of Engineering P.O.Box: 3756-41635,University of Guilan, Rasht, Iran

چکیده:

Image Quality Assessment (IQA) has manyapplications in digital image processing and many techniqueshave been proposed for its measurement such as Mean SquaredError (MSE). In this paper, a new technique, Edge Base ImageQuality Assessments (EBIQA) is proposed. It is based ondifferent edge features which are computed from original anddistorted images. We implement and test the proposed techniqueusing different input samples. Our experimental results showthat the EBIQA technique obtains different values for differentdistorted images which are belong to an original image, whileMSE technique does not. In addition, our proposed technique isconsistent with the mean opinion score that is an IQAmeasurement evaluated by a number of expert observers. Allmeans that the proposed technique is suitable algorithm forautomatic image quality assessment compared to othertechniques.