Validating a quality perception model for image compression: The subjective evaluation of the cogisen’s image compression plug-in

Abstract

User experience has a fundamental role in determining the effectiveness of image compression methods. This work presents the subjective evaluation of a new compression plug-in for current compression formats developed by Cogisen. The quality of image compression methods is often evaluated by objective metrics based on subjective quality datasets, rather than by using subjective quality evaluation tests. Cogisen’s compression method follows an adaptive compression process that evaluates the saliency of any image and calculates the level of compression beyond which viewers shall be aware of image quality degradation. The Single Stimulus Continuous Quality Scale method was used to conduct the subjective quality evaluation of image compression. Pictures compressed by the Facebook Mobile lossy JPEG compression and by the Cogisen plug-in integrated in the Facebook Mobile compression settings were used. The results of the user quality evaluation of pictures show about a 45 % compression improvement, with no loss in perceived image quality, for pictures compressed by the Cogisen plug-in compared to jpeg pictures as compressed by Facebook Mobile.