| 摘要 |
Embodiments of the invention include a method for detecting visual stress scenes in a video. First, a video that includes scenes that may cause visual stress in some viewers is received. Then, a signal representing human perceptual response from scenes in the video is generated. An envelope follower is applied to the perceptual response, from which indications that the video contains visual stress scenes may be generated. The output may also be used to control a contrast limiting circuit, which modifies the original video to remove or minimize those scenes that may cause visual stress. Additionally, output from the envelope follower may be converted to a simulated EEG signal and used to verify accurate modeling by comparing the simulated EEG signal to one measured directly from a viewer watching the same video. |
