遥感图像数据库光谱信息无损压缩方法研究  

Study on Method of Spectral Information in Remote Sensing Image Database

作　　者：李民主[1] 宫纪明[1] LI Min-zhu;GONG Ji-ming(Huaibei Vocational and Technical College,Huaibei 235000,China)

机构地区：[1]淮北职业技术学院,安徽淮北235000

出　　处：《内蒙古民族大学学报（自然科学版）》2020年第3期215-220,共6页Journal of Inner Mongolia Minzu University：Natural Sciences

基　　金：安徽省重点课题项目(2013zjjh049)。

摘　　要：为了减小遥感图像数据库光谱信息的占用空间,需要进行遥感图像数据库光谱信息无损压缩处理,提出一种基于向量量化特征提取的遥感图像数据库光谱信息无损压缩方法,构建遥感图像数据库光谱图像的量化编码模型,采用小波降噪方法进行遥感图像数据库光谱特征的降噪处理,提取遥感图像数据库光谱的关联信息分量,采用信息融合技术进行遥感图像数据库光谱特征的信息融合处理,采用稀疏角点标注方法进行遥感图像数据库光谱信息的关键特征点定位,采用LGB算法实现遥感图像数据库光谱信息无损压缩优化设计.仿真结果表明,采用该方法进行遥感图像数据库光谱信息压缩的信息保真性较好,输出信噪比较高,丢包率较低,提高了信息的存储容量.A lossless compression method for spectral information of remote sensing image database based on vector quantization feature extraction is proposed. A quantization coding model of spectral image of remote sensing image database was constructed. Wavelet denoising method was used to reduce the noise of spectral feature of remote sensing image database,and the correlation information component of spectral spectrum of remote sensing image database was extracted. The spectral feature of remote sensing image database was processed by information fusion technology,and the key feature points of remote sensing image database spectral information were located by sparse corner marking method. The LGB algorithm was used to realize the optimal design of lossless compression of spectral information in remote sensing image database. The simulation results show that the proposed method has good fidelity and the output signal and noise are high,the packet loss rate is low,and the storage capacity of the information is improved.

关 键 词：遥感图像 数据库 光谱信息 无损压缩 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]