On the measurement of image quality perception using frontal EEG analysis

The question of objective measurement of quality of images remains an opened issue. In this paper, we conduced a study concerning the direct measurement of perception of image quality using frontal electroencephalography (EEG). In our work, subjects viewed a series of images for a short predefined period of time while their brain activity was registered using a unique frontal probe EEG instrumentation. The images was organised in 3 classes : sharp, blurry or noisy. Then, by formalizing the task of classification as a supervised learning problem, we have been able correctly identify, on a single-trial basis, the type of images visualized by the user using only their EEG signal. In the context of networked content distributors, this kind of online measurement becomes critical as the objective is to maximize the so-called quality of experience (QoE) of their users while rationalizing their infrastructure utilization.