Parallel image uncompressing system based on multiple processes and multiple threads

Abstract

The invention discloses a parallel image uncompressing system based on multiple processes and multiple threads. The parallel image uncompressing system is used for solving the problem that, in the prior art, the multiple processes are competitive in sharing resources. The parallel image uncompressing system comprises an image front-end processing module, a data sharing memory module, a communication sharing memory module, an image uncompressing module and an image back-end processing module. The image front-end processing module creates uncompressing processes and uncompressing threads, prepares a compressed code stream and starts unoccupied uncompressing threads. The data sharing memory module distributes an independent compressed code stream area and an independent image recovery area to each uncompressing thread. The communication sharing memory module distributes identification areas to all uncompressing threads and is used for conducting communication among the uncompressing threads. The image uncompressing module is used for completing an independent uncompressing process. The image back-end processing module fetches recovery images. The parallel image uncompressing system based on the multiple processes and the multiple threads has the advantages of being high in image uncompressing speed, high in data throughput rate and high in reliability. The parallel image uncompressing system can be used for spatial image uncompressing in deep space communication.

Description

technical field [0001] The invention belongs to the technical field of image processing and relates to image decompression, in particular to a parallel image decompression system involving multiple processes and threads, which can be used for decompression of space images in deep space communication. Background technique [0002] With the rapid development of space image data business, ground image decompression technology is facing the problem of rapidly increasing task load. In order to solve this problem, an efficient parallel image decompression system is needed, which can complete a large number of image decompression tasks with high reliability, high speed and high throughput, which often involves multi-process and multi-thread parallel processing. [0003] There are three ways to complete a large number of image decompression tasks: multi-process and multi-thread, and composite multi-process and multi-line. No matter which method is used, the more processes and threa...

AI Technical Summary

Problems solved by technology

[0005] Shared memory is a free process communication mechanism. All processes can access shared memory. When multiple processes access the same shared memory at the same time, there is competition for shared memory resources because there is no restriction mechanism; while operating the same shared memory , data errors may occur, resulting in abnormal image decompression

For example, two processes write data at the same time, or one process has finished writing data, and the other process has not finished reading data, and other processes start writing data again, resulting in read and write conflicts, making read and write data errors

Images