摘要
高分二号(GF-2)是我国高分辨率对地观测系统重大专项中的首颗"亚米级"卫星,极大地提高了我国高空间分辨卫星影像的自给率。遥感影像融合是高分辨遥感处理和应用的关键流程,选择北京、大连和福州地区GF-2影像,采用彩色标准化变换(Brovey)法、相位恢复(Gram-Schmidt)法、色度空间变换(HSV)法、主成分变换(PCA)法和超分辨率贝叶斯(Pansharp)法共5种融合方法对实验区影像进行融合实验。首先对5种融合结果进行目视效果评价,然后构建了标准差、信息熵、平均梯度、相关系数和扭曲程度5个指标对融合结果进行定量评价。结果表明:Pansharp法在图像信息、细节以及光谱都具有较好的保持效果,Gram-Schmidt和PCA次之,HSV和Brovey色彩失真较明显,该结论可为GF-2卫星影像科研和工程应用提供参考。
GF-2 was the second satellite in major projects of high-resolution earth observation satellite system and it have collected a large number of high resolution satellite images since it was lunched. Image fusion method is the key process in high resolution remote sensing application. Beijing, Dalian and Fuzhou was selected as experi-ment area and five image fusion methods of Brovey, Gram-Schmidt,HSV, PCA and Pansharp were employed to process panchromatic and multispectral imaging. In order to evaluate the result7 performances, firstly, five kinds of fusion results were evaluated by visual, and then established standard deviation, entropy, average gradient, corre-lation coefficient and distortion for quantitative evaluation of fusion results. The results demonstrated that Pansharp was the best algorithm for image information and spectral fidelity of GF-2, followed by Gram-Schmidt and PCA, HSV and Brovey fusion performances were poor. This conclusion can provide key reference for scientific research and engineering application using GF-2 satellite PMS image.
出处
《科学技术与工程》
北大核心
2017年第15期120-125,共6页
Science Technology and Engineering
基金
国家重点研发计划全球变化及应对专项课题(2016YFA0600302)
海南省重大科技计划项目(ZDKJ2016021)和海南省重大科技计划项目(ZDKJ2016015-1)资助
