基于多核DSP的星载并行遥感图像压缩系统设计与实现

发布时间：2018-03-28 01:26

本文选题：图像压缩　切入点：JPEG　出处：《计算机应用》2017年05期

【摘要】：随着星载遥感技术的不断发展,产生的遥感数据也变得日渐庞大,目前有限的通信带宽远不能满足遥感图像数据传输的需求。因此研究面向星载应用的图像压缩技术对空间应用技术的发展有着十分重要的意义。采用传统单核数字信号处理器(DSP)难以满足性能需求,而采用现场可编程门阵列(FPGA)则难以满足功耗需求,近年来随着硬件技术发展,多核DSP技术已经成熟,且在弹载场景已有比较成熟的多核DSP图像压缩解决方案,可供星载应用参考。基于多核DSP,即TI公司的C6678多核浮点DSP平台,构建一个并行图像压缩系统,并充分利用多核DSP的硬件资源。考虑星载遥感图像压缩对压缩质量、压缩速度等多方面指标都有着较高的要求,系统采用JPEG2000标准进行图像压缩,并且采用了主核负责外部通信与内部任务分配、从核执行JPEG2000图像压缩的设计方案。测试结果表明,该系统运行稳定可靠,且整体压缩性能优秀,能够满足对星载遥感图像压缩系统的性能要求。
[Abstract]:With the continuous development of space-borne remote sensing technology, the resulting remote sensing data is becoming larger and larger. At present, the limited communication bandwidth is far from meeting the demand of remote sensing image data transmission. Therefore, it is very important for the development of space application technology to study the image compression technology for space-borne applications. Word signal processor (DSP) is difficult to meet performance requirements, In recent years, with the development of hardware technology, multi-core DSP technology has matured, and there is a mature multi-core DSP image compression solution in missile-borne scene. It can be used as a reference for spaceborne applications. A parallel image compression system is constructed on the basis of multi-core DSP, that is, C6678 multi-core floating-point DSP platform of TI, and the hardware resources of multi-core DSP are fully utilized. The compression quality of space-borne remote sensing image compression is considered. The system adopts JPEG2000 standard for image compression, and adopts the main core responsible for external communication and internal task allocation, and executes the design scheme of JPEG2000 image compression from the core. The test results show that, The system runs stably and reliably, and the overall compression performance is excellent, which can meet the performance requirements of spaceborne remote sensing image compression system.
【作者单位】：国防科学技术大学计算机学院;
【基金】：国家自然科学基金资助项目(61272146) 国防科学技术大学优秀研究生创新资助项目(B151101)~~
【分类号】：TP751

【相似文献】