摘要
针对高分二号(GF-2)卫星全色遥感影像与哨兵二号(Sentinel-2)卫星多光谱遥感影像空间分辨率相差较大且传感器不同导致的光谱或空间信息丢失问题,结合快速离散Curvelet变换对HSV融合方法的分量替换过程进行改进,利用GF-2的高空间分辨率和Sentinel-2的多光谱特性分别设计高、低频系数融合规则,并且与4种经典算法的融合结果进行比较研究。通过目视解译对融合影像进行初步定性评价,然后计算熵值、平均梯度、偏差度、影像相关系数评价、均值、空间频率、结构相似性、峰值信噪比、光谱扭曲度等10个评价指标对融合影像进行定量评价。通过与4种经典算法融合对比结果可知,改进的HSV融合方法能够最大程度保留原始全色影像的空间信息与多光谱影像的光谱信息。
The focus of this paper is on the fusion problem encountered by satellites such as GF-2 panchromatic remote sensing image and Sentinel-2 multispectral remote sensing image.Through a combination of fast discrete Curvelet transformations,an improvement on the vector switch process of the HSV fusion method is achieve.We use the high spatial resolution of GF-2 and the multispectral characteristics of Sentinel-2 to design the fusion rules of high and low frequency coefficients,and compare the results with the fusion results of four classic algorithm;and using preliminary qualitative evaluation criteria,such as visual interpretation,and quantitative evaluation criteria such as,entropy,average gradient and deviation,image correlation coefficient evaluation criteria,mean value,spatial frequency,structural similarity,peak signal-to-noise ratio and spectral distortion.Compared with the fusion results of four classic algorithms,it shows that the improved HSV fusion method has the best effect and has managed to keep the original panchromatic and multispectral spatial information of the images intact.
作者
杨军
王筱宇
YANG Jun;WANG Xiaoyu(Faculty of Geomatics,Lanzhou Jiaotong University,Lanzhou 730070,China;National-Local Joint Engineering Research Center of Technologies and Applications for National Geographic State Monitoring,Lanzhou 730070,China;Gansu Provincial Engineering Laboratory for National Geographic State Monitoring,Lanzhou 730070,China)
出处
《测绘科学》
CSCD
北大核心
2022年第1期112-120,共9页
Science of Surveying and Mapping
基金
国家自然科学基金项目(61862039)
甘肃省科技计划资助项目(20JR5RA429)
兰州市人才创新创业项目(2020-RC-22)
兰州交通大学天佑创新团队项目(TY202002)。
