高光谱图像的稀疏表示和压缩算法研究
发布时间：2019-07-09 07:49
【摘要】：高光谱遥感技术是20世纪末发展起来的，是融合电磁学、光学、信号处理等多学科交叉领域的新兴学科。与传统遥感技术相比，高光谱遥感技术在获取地面信息的同时，还可获取丰富的地物光谱信息，使其在农业、林业、地质、环境、军事等不同领域得到了广泛的应用。随着空间分辨力和谱间分辨力的不断提高，高光谱遥感图像的数据量呈指数量级增长，给传输和存储带来了巨大的压力。因此研究高光谱图像压缩算法对高光谱遥感技术的发展有着至关重要的意义。 本文针对高光谱遥感图像在实际应用中面临的数据量庞大，信息获取和数据传输之间的矛盾日益加剧等一系列问题，对基于冗余字典的高光谱遥感图像的稀疏表示和压缩算法进行了深入研究。主要研究内容如下： (1)实现了基于冗余字典的高光谱遥感图像的稀疏表示。该方法能够以较少的数据量更好地描述高光谱图像中的特征信息，是一种有效的高光谱图像表示方法。 (2)研究了一种基于稀疏表示的高光谱遥感图像的压缩方法。该方法在对高光谱遥感图像进行稀疏表示的情况下，，采用比特平面编码对稀疏表示系数进行压缩，获得了较高的压缩比。 (3)完成了对高光谱遥感图像的重建，并获得了良好的重建效果。 本文进行了大量仿真实验，实验结果表明本文算法能够取得良好的压缩效果和良好的重构效果，在不同的高光谱图像库中具有较好的通用性。
文内图片：
图片说明：本文系统框架
[Abstract]:Hyperspectral remote sensing technology was developed at the end of the 20th century. It is a new subject which integrates electromagnetics, optics, signal processing and other interdisciplinary fields. Compared with traditional remote sensing technology, hyperspectral remote sensing technology can not only obtain ground information, but also obtain rich spectral information of ground objects, which has been widely used in agriculture, forestry, geology, environment, military and other fields. With the continuous improvement of spatial resolution and inter-spectral resolution, the amount of data of hyperspectral remote sensing images increases in the order of magnitude, which brings great pressure to transmission and storage. Therefore, it is of great significance to study hyperspectral image compression algorithm for the development of hyperspectral remote sensing technology. In order to solve a series of problems, such as huge amount of data and increasing contradiction between information acquisition and data transmission, the sparse representation and compression algorithm of hyperspectral remote sensing image based on redundant dictionary is deeply studied in this paper. The main research contents are as follows: (1) the sparse representation of hyperspectral remote sensing images based on redundant dictionaries is realized. This method can better describe the feature information in hyperspectral images with less data, and it is an effective hyperspectral image representation method. (2) A compression method of hyperspectral remote sensing image based on sparse representation is studied. In this method, the sparse representation coefficient is compressed by bit plane coding under the condition of sparse representation of hyperspectral remote sensing images, and a high compression ratio is obtained. (3) the reconstruction of hyperspectral remote sensing image is completed, and good reconstruction effect is obtained. In this paper, a large number of simulation experiments are carried out, and the experimental results show that the algorithm can achieve good compression effect and good reconstruction effect, and has good generality in different hyperspectral image libraries.
【学位授予单位】：河北大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP751
本文编号：2511991
文内图片：
图片说明：本文系统框架
[Abstract]:Hyperspectral remote sensing technology was developed at the end of the 20th century. It is a new subject which integrates electromagnetics, optics, signal processing and other interdisciplinary fields. Compared with traditional remote sensing technology, hyperspectral remote sensing technology can not only obtain ground information, but also obtain rich spectral information of ground objects, which has been widely used in agriculture, forestry, geology, environment, military and other fields. With the continuous improvement of spatial resolution and inter-spectral resolution, the amount of data of hyperspectral remote sensing images increases in the order of magnitude, which brings great pressure to transmission and storage. Therefore, it is of great significance to study hyperspectral image compression algorithm for the development of hyperspectral remote sensing technology. In order to solve a series of problems, such as huge amount of data and increasing contradiction between information acquisition and data transmission, the sparse representation and compression algorithm of hyperspectral remote sensing image based on redundant dictionary is deeply studied in this paper. The main research contents are as follows: (1) the sparse representation of hyperspectral remote sensing images based on redundant dictionaries is realized. This method can better describe the feature information in hyperspectral images with less data, and it is an effective hyperspectral image representation method. (2) A compression method of hyperspectral remote sensing image based on sparse representation is studied. In this method, the sparse representation coefficient is compressed by bit plane coding under the condition of sparse representation of hyperspectral remote sensing images, and a high compression ratio is obtained. (3) the reconstruction of hyperspectral remote sensing image is completed, and good reconstruction effect is obtained. In this paper, a large number of simulation experiments are carried out, and the experimental results show that the algorithm can achieve good compression effect and good reconstruction effect, and has good generality in different hyperspectral image libraries.
【学位授予单位】：河北大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP751
【参考文献】
相关期刊论文 前10条
1 张宇,尹昊晖,张家谋;图象质量客观测试的研究[J];北京邮电大学学报;1999年04期
2 杨国鹏;余旭初;冯伍法;刘伟;陈伟;;高光谱遥感技术的发展与应用现状[J];测绘通报;2008年10期
3 夏豪;张荣;;基于改进预测树的超光谱遥感图像无损压缩方法[J];电子与信息学报;2009年04期
4 刘恒殊,彭风华,黄廉卿;超光谱遥感图像特征分析[J];光学精密工程;2001年04期
5 孙蕾;罗建书;谷德峰;;基于谱间预测和码流预分配的高光谱图像压缩算法[J];光学精密工程;2008年04期
6 王继林;;比特平面编码用于图像压缩的程序设计[J];电脑编程技巧与维护;2008年06期
7 汪孔桥;数字图像的质量评价[J];测控技术;2000年05期
8 肖竹;王素玉;卓力;;成像光谱图像压缩技术研究的新进展[J];测控技术;2009年05期
9 张春梅;尹忠科;肖明霞;;基于冗余字典的信号超完备表示与稀疏分解[J];科学通报;2006年06期
10 刘丹华;石光明;周佳社;;一种冗余字典下的信号稀疏分解新方法[J];西安电子科技大学学报;2008年02期
本文编号：2511991
本文链接：https://www.wllwen.com/guanlilunwen/gongchengguanli/2511991.html
