摘要
遥感图像压缩会影响分类精度,是值得研究的问题。以高分辨率遥感影像(Quick Bird)的监督分类精度评定为尺度,采用ER Mapper软件的JPEG 2000图像压缩模块对图像进行压缩,再在eCognition软件中对这9种压缩比图像进行面向对象的监督分类,生成分类精度报告。通过分析分类精度的变化,研究了JPEG 2000压缩对遥感影像分类的影响程度及其在遥感影像压缩方面的应用潜力。
How remote sonsing image compression affects the classification accuracy is a hot question. This paper makes some experimental investigations into the effect of JPEG 2000 compression of ER Mapper software on the accuracy of supervised classification of multispectral QuickBird image. The object-oriented supervised classification software, eConigition, is used for classification with 9 compression ratio images, and the report of classification accuracy is generated. By analyzing the changes of classification accuracy, it discnsses the effect of JPEG 2000 on the classification accuracy and the potential of this compression method in the field of remote sensing.
出处
《测绘工程》
CSCD
2007年第1期23-26,共4页
Engineering of Surveying and Mapping
基金
湖南省科技计划资助项目(2004022200611)
关键词
图像压缩
JPEG
2000
监督分类
分类精度
精度评定
image compression
JPEG2000
supervised classification
classification accuracy
accuracy assessment
