Realization method of AVS (Audio Video Standard) inverse transformation based on reconfiguration technology

Abstract

The invention relates to a realization method of AVS (Audio Video Standard) inverse transformation based on a reconfiguration technology, and the method comprises the following steps of: firstly, carrying out algorithm analysis, namely analyzing a C code of an algorithm and extracting a DFG (Data Flow Graph) of the algorithm to obtain a data transmission requirement of the algorithm; secondly, dividing and mapping the DFG according results of the algorithm analysis and a frame of a reconfigurable array, so as to design the optimal data transmission scheme; thirdly, utilizing a configuration tool to generate configuration characters of the reconfigurable array according to the results of the two previous steps; and finally, loading configuration information into a configuration information memorizer of the reconfigurable array through an ARM (Advanced RISC Machine) processor, so as to reconfigure the reconfigurable array into an acceleration module which is specially used for executing the inverse transformation. According to the realization method provided by the invention, a complicated chip design process is not needed, so that the development time and the development cost can be greatly saved and the practical applicability is very high.

Description

technical field [0001] The present invention relates to a method in the field of embedded video decoding, in particular to an implementation method of AVS inverse transformation based on reconfigurable technology. Background technique [0002] With the development of video standards, the effect and performance of video compression are getting better and better, but its complexity and calculation amount are also greatly increased. Correspondingly, to realize real-time decoding at the decoding end, high requirements are put forward for the data parallelism and computing efficiency of the hardware. [0003] Standard (Audio and Video Coding Standard) is a second-generation source coding standard proposed by my country with independent intellectual property rights, and its coding efficiency is equivalent to that of H.264. Its video part officially became a national standard in 2006. [0004] Transform coding is an important method to remove the spatial correlation of pixels and...

AI Technical Summary

Problems solved by technology

However, as the scale and complexity of the algorithm increase, the number of FPGA units and the interconnection area increase sharply, and the power consumption increases at the same time.

Images