JPEG 2000 speed control method through real time cut-off

Abstract

Disclosed method predicts effective bit plane in a internal code block within sub band, and puts forward a scheme for controlling preallocation rate of code block and code stream. Wavelet weight omega and quantization coefficient delta are as weights in allocating code rates between code blocks, and rate of block is allocated according to total number of channel contained in code block. Based on the algorithm said above, VLSI structure in real time coding and real time blocking is put forward, and JPEG 2000 does this requirement, according to the experiment proof. The invention raises coding speed and compression ratio in duplication so as to solve bottleneck in image video compression.

Description

technical field [0001] The invention belongs to the technical field of VLSI design. Specifically, it involves designing a real-time truncated JPEG2000 rate control method in JPEG2000 hardware implementation. Background technique [0002] In actual systems such as digital cameras, videophones, camcorders, and handheld computers, considering the requirements of processing speed and volume, it is an inevitable choice to use chips to achieve image data compression. In the currently known JPEG2000 chip structure design, if the compression ratio of image data is very low, the advantages of JPEG2000 cannot be reflected. When the compression ratio is relatively high, rate control must be performed to truncate the coded stream. . [0003] JPEG2000 rate control performs optimal hierarchical truncation on the code stream after arithmetic coding, that is, truncates the code stream according to the compression ratio set by the user, so as to meet the reserved code stream and achieve be...

AI Technical Summary

Problems solved by technology

In addition, the PCRD optimization algorithm is a global optimization process on the tile block. It must wait until all the channels of all code blocks in the entire Tile block are encoded before performing a search for the hierarchical truncation point, which requires a large amount of memory to Store the truncated code stream and the RD slope of each encoding channel, and will cause the real-time performance of the entire encoding to decrease

In addition, the PCRD optimization algorithm needs to search for hierarchical truncation points. This search process needs to traverse every encoding channel on the entire tile block every time, and it takes many times to complete the search. In the case of a large number of layers, the search will cost plenty of time

The PCRD optimization algorithm has the shortcomings of large memory overhead and poor real-time performance of the system. It is necessary to further optimize the JPEG2000 compression rate control algorithm from the perspective of hardware implementation or software implementation.

Images