Due to technology advances in multimedia, larger storage spaces, large internet bandwidth and high-transmission speed are required for transmission videos. Video compression techniques play a vital role in reducing video size, therefore, smaller storage space and lower internet bandwidth are eventually required. In this paper, the EEG signal is used to modify the compression ratio of videos based on the interest of the viewer. This is performed by associating the compression ratio applied to the video with the degree of interest using a group of frames. This interest for a group of frames is measured using the EEG signal, to demonstrate the viewer responses to videos. Statistical techniques applied to the EEG signal (such as peaks-over-threshold and time-of-peaks over-thresholds) are used to extract the frames of interest. Peak signal-to-noise ratio (PSNR), Structural Similarity Index (SSIM) and the Mean-Square Error (MSE) are used to compare the performance of the proposed technique with the MPEG-4 technique. The results show a reduction of 15 % in the video size compared with the MPEG-4 technique without deteriorating the quality of the videos.
Key words: Video Compression, EEG signal, MPEG4, PSNR, SSIM, Brainwaves.
|
