文档介绍：西安电子科技大学
硕士学位论文
星载无损图像压缩系统设计与实现
姓名:郭瑞
申请学位级别:硕士
专业:通信与信息系统
指导教师:李云松
20100101
摘要
摘要
随着航天遥感技术的迅速发展,各类图像载荷被搭载在同一颗卫星上。为了
能够在降低数据率的同时保证卫星获取的宝贵数据能够最大限度地满足地面各类
用户处理的需求,往往需要对这些数据进行无损和近无损压缩,尤其对于以科学
应用为目的的数据如多光谱图像、SAR 成像数据等一般要求采用无损压缩。由于
卫星资源十分宝贵,对有效载荷的功耗、体积、重量以及可靠性等指标有很高的
要求,因此研究低存储、低功耗和具有高可靠性的实时图像无损编码器在航天领
域具有很重要的意义。
JPEG-LS 算法采用预测与 Golomb 编码,在已有的图像无损压缩算法中,
JPEG-LS 算法是公认压缩性能最高的算法之一,其无损压缩性能优于 JPEG2000 算
法。同时JPEG-LS算法复杂度较低,易于硬件实现,因此本文在仔细研究了JPEG-LS
图像压缩算法的基础上,重点研究该算法的 FPGA 硬件实现方法。
本文深入研究了 JPEG-LS 图像压缩算法原理及实现,设计了 JPEG-LS 算法的
高效 VLSI 结构,并采用 FPGA 加以实现,其中提出一种新型桶型移位器的实现方
法,该方法具有更高的效率。本文还根据某星载设备实际应用的要求,将设计好
的 JPEG-LS 模块作为一个 IP 核嵌入到外围接口中,最终设计并实现了完整的图像
无损压缩系统。经过大量测试,该系统工作稳定,各项技术指标达到了既定要求。

关键词:JPEG-LS Golomb 编码 VLSI 无损压缩 FPGA
Abstract
Abstract
With the rapid development of space remote sensing technology, all kinds of
images in the same payload were equipped on a satellite. Lossless pression is
often needed, in order to reduce the data rate while ensuring the satellite to obtain
valuable data which can satisfy the requirements of ground handling of various types of
users, especially for multi-spectral image, SAR image data, etc. which aim at scientific
applications. As the valuable satellite resources have high expectations on, right payload
power, volume, weight, reliability and other indicators, the research on real-time image
lossless encoder, which has low storage, low power consumption, high reliability, and
high-performance, is of critical significance in the field of aerospace study.
JPEG-LS algorithm, which is one of the best existing lossless pression
algorithms with superior performance, adopts prediction and Golomb coding. Its
performance outperforms JPEG2000 pression algorithm. Meanwhile,
JPEG-LS algorithm has plexity, and is easier for hardware implementation.
Hence, this paper focuses on the research of FPGA hardware implementation of
JPEG-LS pression algorithm.
This paper first presents a in-dep