| 摘要 |
An encoder including a DCT transformator for receiving an image or video frame and performing DCT transformation upon a pixel block in a macroblock of the frame, thereby providing a matrix of DCT coefficients, a zig-zag matrix-to-vector converter for receiving the matrix of DCT coefficients from the DCT transformator and creating a one-dimensional vector of DCT coefficients, an analyzer for receiving the one-dimensional vector of DCT coefficients and classifying the macroblock according to a level of detail and edginess, a rate controller for receiving the classification information from the analyzer and selecting DCT filtering parameters, a DCT filter for receiving the DCT filtering parameters selection from the rate controller and implementing the DCT filtering parameters upon the frame, thereby providing a filtered DCT coefficient vector, a quantizer for quantizing the filtered DCT coefficient vector, and a compressor for compressing the quantized results.
|
