Quality Assessment of Post-Processed Images

摘要

The vast majority of the work done in the field of quality assessmentduring last two decades has been dedicated to the quantification ofthe distortion caused by the processing of an image. The originalimage was, therefore, always considered to be of the best possiblequality. In this kind of scenario, the notion of quality can beexpressed as the fidelity of the processed version to the reference.However, some post-processing algorithms enable to adjustaesthetic properties of an image in order to enhance the perceivedquality. In such cases, the best possible quality image is notavailable and the classical fidelity approach is no longer applicable.The goal of this thesis is to revise the quality assessmentmethodologies to cope with the challenges brought by thepost-processing into the quality evaluation. The post-processingalgorithms, relevant to the topic of this thesis, come from two groups– image enhancement, represented by image sharpening, anddynamic range compression (also known as tone-mapping)techniques. Both subjective and objective quality assessmentmethodologies applicable in these areas are studied and thesuitable solutions, outperforming the state-of-the-art methods, areproposed. Moreover, a novel methodology for evaluating theperformance of objective quality metrics, overcoming theshortcomings of the currently available methods, is presented.