Quantized matrix adjusting method for improving quality of video

Abstract

It is a kind of quantification matrix adjusting method that can improve the image quality. It comprises the following steps: updating the used bit amount and average quantification ratio factor; appointing the bit precalculating value of the current image; calculating and estimating the quantification ratio factor according to the used bit amount, average quantification ratio factor and bit precalculating value; judging that whether enters the adjustment mode: when the estimated quantification ratio factor less than the threshold value of quantification ratio factor, the encoding mode is set as adjustment mode, otherwise it is general mode; adjusting the Quantize Matrix: when the encoding mode is set as adjustment mode, it reduces the predetermined Quantize Matrix; encoding: if the encoding mode is set as adjustment mode, encoding based on the reduced Quantize Matrix, if the encoding mode is general mode, encoding based on the predetermined Quantize Matrix. For the reduction of predetermined Quantize Matrix in the adjustment mode, it can avoid the underestimation of quantification ratio factor, at the same time boost the image quality.

Description

technical field [0001] The invention relates to a quantization matrix adjustment method for improving image quality. Background technique [0002] In order to reduce the amount of video data, generally the video is compressed, such as video coding in MPEG format. The basic unit of image coding is a macro-block. If the sampling format is 4:2:0, each macroblock has 6 blocks, namely 4 Y blocks, one Cb block, and one Cr block. Each block must first undergo a discrete cosine transform (hereinafter referred to as DCT), and then the DCT coefficients (coefficients) are quantized into integers (integers). A zigzag scan or other alternative scan method is used to arrange the two-dimensional array of quantized coefficients into one-dimensional data. Finally, so-called variable-length coding (variable-length coding, hereinafter referred to as VLC) is used to perform entropy coding. [0003] In general, the quantization of DCT coefficients F[v][u] can be expressed as: [0004] QF[v]...

AI Technical Summary

Problems solved by technology

However, when the sequence of video is simple and the target bit rate is high, the quantization scale factor can be made very small to produce more bits

Even when the video is really simple, even if the quantization scale factor Q is set to 1, it cannot generate enough bits, so that it needs to increase the stuffing bits, which wastes the bit rate

In addition, when the average quantization scale factor Avg_Q is 1.5, the most appropriate quantization scale factor for most macroblocks should be 1.5, but 1 or 2 must be selected

Images