The Efficacy of Frontal and Temporal Lobes in Detecting 2D&3D Video Transition Using EEG Power

Abstract

Due to the three-dimensional (3D) structure of the human eye, 3D technology was used in this research. Transition to 2D and 3D analysis is important, claiming that binocular vision will lose dimension during fatigue. Thus, a single-stream video consisting of random 2D&3D parts was watched by nine volunteers. The dynamic behavior and power spectral density (PSD) of the volunteers’ brain signals were considered for a comprehensive quantitative analysis. The dominant EEG bands and time intervals were selected in 2D to 3D (2D_3D) and 3D to 2D (3D_2D) transitions by analyzing power differences based on short-time Fourier transformation (STFT). Taking into account this information, applying effective feature extraction and classification techniques, the behavioral patterns of channels representing the brain lobes of the volunteers were analyzed. Hjorth parameters and statistical methods were used as feature extraction methods. The k-nearest neighbors (k-NN) and linear discriminant analysis (LDA) algorithms were applied to classify 2D_3D and 3D_2D transitions. The results revealed that, thanks to the comprehensive classification analysis of the 2D_3D and 3D_2D transitions, the change in the activity power of the brain cortex can be represented. The dominance of the temporal and frontal lobes between the channels representing these transitions cannot be excluded.