Image distortion threshold coefficient estimation method based on EEG (electroencephalogram) signals

Abstract

The invention discloses an image distortion threshold coefficient estimation method based on EEG (electroencephalogram) signals, and mainly solves the problem of low accuracy rate of image quality evaluation due to the fact that human vision characteristics are not fully considered in the prior art. According to the scheme of the method, stimulating images of different distortion levels are designed; the EEG signals of a subject observing the stimulating images are collected; the collected EEG signals are sequentially subjected to reference conversion, baseline correction, filtering and segment processing; an EEG signal threshold with the accuracy rate of 50% is calculated according to the processed EEG signals when the subject observes the distorted images, a linear relation between the EEG signal threshold of the subject and the image distortion levels is obtained, and an image distortion threshold coefficient is estimated. According to the method, the human vision perception characteristics are fully considered, the EEG signal threshold caused when the subject observes the distorted images is estimated according to the image distortion threshold coefficient, objectivity and accuracy of image quality evaluation are improved, and the method can be applied to image quality evaluation, image compression and image detection.

Description

technical field [0001] The invention belongs to the technical field of image processing, and further relates to an image distortion threshold coefficient estimation method. It can be used for image quality evaluation, image compression, image detection and psychological behavior research. Background technique [0002] With the continuous development of display technology, consumers in modern society have higher and higher requirements for display technology. Not only do they require basic things such as large screens, clear images, and rich colors, but consumers begin to care more about deeper things, such as Whether it is true when watching the image, and whether it is the image quality you most hope to achieve. Because of this, the research on image quality evaluation has become a hot spot in display technology research. [0003] Image quality evaluation methods are divided into subjective evaluation and objective evaluation. Subjective evaluation uses people as observe...

AI Technical Summary

Problems solved by technology

[0007] The technical solution of the present invention is, aiming at the limitations and uncertainties existing in the existing image quality evaluation methods, resulting in insufficient consideration of the human visual characteristics, by designing the research paradigm of EEG experiments, extracting the subjects’ EEG Signal, associating EEG signals with behavioral data, estimating the perceptible thresholds of EEG signals of images with different degrees of distortion, and obtaining the relationship between the degree of image distortion and the perceivable thresholds, thereby obtaining the image distortion threshold coefficients. The implementation steps include the following:

Images