基于DCT和DWT的遥感图像压缩算法比较  

Comparison of Algorithms Based on DCT and DWT in Remote Sensing Image Compression

作　　者：严俊雄[1] 王文[1] 李子扬[2] 李安[1] 陈勃[1] 

机构地区：[1]中国科学院对地观测与数字地球科学中心,北京100086 [2]中国科学院研究生院

出　　处：《科学技术与工程》2008年第19期5439-5445,共7页Science Technology and Engineering

基　　金：中国科学院知识创新工程青年人才领域前沿项目资助

摘　　要：由于遥感图像具有数据量大、分辨率高、覆盖范围广、纹理复杂、细节多、灰度变化大、目标小、空间相关性较差等特点,很难对其进行高比例压缩。因此遥感图像的及时显示、共享与数据传输问题,一直是遥感数据存储、处理与分析过程中存在的技术瓶颈。另一方面,在目前网络带宽有限的情况下,这也为GIS(地理信息系统)信息的实时动态应用造成了困难。因此,图像压缩在遥感数据存储、传输与共享等应用上有很重要的作用。探讨应用在遥感领域的DCT(JPEG)和DWT(JPEG2000,ECW,MRSID)算法,然后采用基于这些算法的技术工具比较压缩重建后图像的质量。最后,针对不同的遥感应用提出了一些建议。Due to the special characteristics such as large size, high-resolution, wide coverage, small targets, complex texture, more details, and poor spatial correlation, remote sensing image can not be easily compressed at a high ratio. It is difficult to share and transmit the image data in time as well as display. This is so-called ＂technological bottleneck＂ existing in data archiving and data analysis. On the other hand, the network bandwidth is too limited to achieve a dynamic and real-time GIS Information application. So image compression plays an increasingly important role in remote sensing data share. This paper mainly discusses the algorithms and techniques based on DCT （JPEG） and DWT （JPEG2000, ECW and MRSID） in remote sensing image compression, and then compare the quality of compressed images by using these algorithms or techniques. Finally, suggestion is given to select a suitable algorithm or technique for remote sensing data applications.

关 键 词：DCT DWT 遥感图像 图像压缩 

分 类 号：TP751.1[自动化与计算机技术—检测技术与自动化装置]