| 摘要 |
A method for compressing multiple channels of data using an MPEG-compatible motion estimation technique. The method determines the differences between frames within each channel and among the channels and encodes those differences. To facilitate compression, some of the frames are predicted and some of the frames are skipped. The differences between the current frame and the predicted frame for each channel are computed for each successive group of frames, and then compared, after each group of frames of each channel is encoded, to a threshold level E0. In the event that, for any of the groups of frames, the difference between the current frame and the predicted frame of any of the channels reaches the threshold, the method of the present system will insert an I-frame into the sequence of encoded frames, thus restarting the encoding process with an I-frame. The method of computing differences and encoding I-frames is implemented for both intra-channel differences (within each channel) and inter-channel differences (between the channels). The encoded frames can thereafter be transmitted to a decoder for reconstructing the frames for display. Preferably, both the encoding and/or decoding operations are implemented with artificial intelligence (AI) or fuzzy logic software.
|
