Study Of The Display Image Quality And Visual Comfort Assessment Based On Multimodal Technology

In recent years,with the rapid development of display technology,people have higher requirements for image quality and visual comfort of display system.How to improve the image quality and visual comfort of display system has become a common concern of researchers and manufacturers.The study reveals the mechanism of how the display system characteristics affecting image quality and visual comfort based on visual attention characteristics and display system characteristics.The image quality and visual comfort have been systematically studied in depth.The corresponding quantitative evaluation methods are proposed to provide a basis for improving the image quality and the performance of the display device.The contents of this paper are as follows:A no-reference video quality evaluation model based on visual attention characteristics is established.This paper proposes a feature integration method based on time domain,central weighted spatial domain and frequency domain saliency model.Objective saliency detection model was built by simulating human eye's visual attention characteristics,color opposition characteristics,visual masking characteristics,central preference characteristics,and motion information.Combined with the subjective quality evaluation of the viewers,a no-reference video quality evaluation model based on the perceptual blur and block metric is established.The model has higher prediction accuracy and lower computational complexity.A blurred image quality evaluation model based on EEG signals is proposed.The human cognitive mechanism of image quality evaluation was investigated using the combination of subjective evaluation and event-related potentials(ERPs)measurement.By analyzing the changes of EEG signals in different scalp areas,detection rate,reaction time and subjective score,the image quality evaluation model is established.The result shows a good correlation with the subjective rating scores,which proved the feasibility of utilizing ERPs to evaluate the blurred image quality.In addition,EEG signals in the anterior middle(AM)scalp area are the most correlated area with image quality.An image quality evaluation model based on fusion of EEG and eye movement signals is proposed.The paper uses subjective evaluation,EEG measurement and eye movement measurement to further study the internal perception and cognitive process of image quality evaluation.By analyzing the central nervous system of EEG measurement and the ocular nervous system of eye movement measurement,an image quality evaluation model based on the fusion of EEG of the AM area and eye movement signals is established.The result has a good correlation with subjective rating scores,and it improves the performance of image quality model.A three-dimensional display visual comfort prediction model based on multimodal technology is established.The paper uses multi-modal parameters fusion method such as ophthalmological parameters,Electrocardiograph(ECG)signals,EEG signals,eye movement signals and subjective questionnaires.By analyzing the change of multimodal parameters before and after watching 70 minutes 2D,3D or 3D film,a visual comfort prediction/evaluation model based on the BP neural network is established.The prediction results have higher precision and the same trend as the verification experiment.