一种GF-2全色多光谱影像融合方法  

A Fusion Method for GF-2 Panchromatic Multispectral Images

作　　者：王筱宇 杨军 WANG Xiaoyu;YANG Jun(School of Science and Technology,Xinyang University,Xinyang 464000,China;Faculty of Geomatics,Lanzhou Jiaotong University,Lanzhou 730070,China)

机构地区：[1]信阳学院理工学院,河南信阳464000 [2]兰州交通大学测绘与地理信息学院,甘肃兰州730070

出　　处：《现代信息科技》2024年第20期107-110,116,共5页Modern Information Technology

基　　金：国家自然科学基金(61462059,61862039);信阳学院科研项目(2023—XJLYB-009)。

摘　　要：高分2号卫星影像的应用日益多样化,如何获取高质量融合影像成为需要研究的重要问题。为了从原始的全色影像和多光谱影像中获取高空间分辨率和高光谱分辨率的影像,结合方向信息和脉冲耦合神经网络对非下采样轮廓波变换算法进行改进。以甘肃省兰州市的GF-2全色影像和多光谱影像作为数据源,提出了一种结合PCNN和NSCT的遥感影像融合方法,通过定性评估和定量评估,与IHS方法、BT方法、PCA方法和GS方法相比,改进后的方法在改善空间细节和保留光谱信息方面具有更好的效果。The application of GF-2 satellite images is becoming increasingly diversified,and how to obtain high-quality fusion images has become an important issue that needs to be studied.In order to obtain high spatial resolution and high spectral resolution images from the original panchromatic images and multispectral images,the Nonsubsampled Contourlet Transform algorithm is improved by combining orientation information and Pulse Coupled Neural Network.Based on the GF-2 panchromatic images and multispectral images of Lanzhou,Gansu Province as the data source,a remote sensing image fusion method combining PCNN and NSCT is proposed.Through qualitative evaluation and quantitative evaluation,compared with IHS method,BT method,PCA method and GS method,the improved method has better effect in improving spatial details and preserving spectral information.

关 键 词：高分2号影像 遥感影像融合 脉冲耦合神经网络 非下采样轮廓波变换 方向信息 

分 类 号：TP751[自动化与计算机技术—检测技术与自动化装置]