Data packaging device and data packaging method

Abstract

The invention provides a data packaging device for continuously receiving current data, comprising N barrel-shaped shifters, a control module and a packaging module. The control module is used for determining a shifting amount based on length of previous residual data, and controlling the N barrel-shaped shifters to orderly apply M-time and barrel-shaped shift on current merging data, in order to achieve the shift amount. The packaging module is used for merging the previous residual data and the current merging data through the N barrel-shaped shifters. Each shifting amount of the M-time barrel shift is smaller than a threshold value, which is relative to a periodic length; and the current merging data is part or all of the current data. N is an integer more than or equal to 2, and M is a natural number less than or equal to N.

Description

technical field [0001] The present invention relates to memory management techniques, and more particularly to memory management techniques for packing data. Background technique [0002] In recent years, many electronic products, including digital cameras, video recorders, smart phones, etc., all attract consumers with the ability to shoot large-size, high-resolution images / videos. For hardware manufacturers, how to keep pace with the times and produce higher-quality video encoders has always been an important issue. [0003] figure 1 An example block diagram of an image encoder. The pixel block input module 11 forms the pixel data input from the data supply end into blocks. The conversion / quantization module 12 applies discrete cosine transform (discretecosine transform, DCT) and quantization procedures to the block generated by the pixel block input module 11 to achieve a compression effect. Subsequently, the run length coding (runlengthcoding, RLC) module 13 is respo...

AI Technical Summary

Problems solved by technology

In the case of a large amount of data, if the bit stream packaging module 15 cannot cooperate with the speed at which the variable length encoding module 14 produces the bit stream, the receiving, shifting, merging, and output of the bit stream can be successfully completed, and the bit stream packaging Module 15 is likely to be the bottleneck that slows down the overall speed of the image encoder

[0009] If the amount of subsequent data is reduced by increasing the degree of quantization used by the conversion / quantization module 12, although the speed requirement for the bitstream packaging module 15 can be alleviated, the image quality will be sacrificed, which is not an ideal solution.

Images