摘要
针对沿海牡蛎养殖模式的特点,使用WorldView-2影像为数据源,以浙江省象山港牡蛎养殖区为研究区,采用主成分分析(Principal Component Analysis,PCA)、GS(Gram-Schmidt)变换、NNDiffuse Pan Sharpening、Brovey变换、小波变换(Wavelet Transform)5种融合方法对多光谱和全色影像数据进行融合,选用均值、标准差、信息熵、平均梯度、相关系数和光谱扭曲程度6种客观评价指标,对5种融合结果进行主观定性和客观定量的评价与分析。结果表明:整体上,经PCA方法融合后的遥感影像在保持空间纹理细节信息的同时,光谱信息保持较好,是WorldView-2影像进行沿海牡蛎养殖遥感应用时最适合的融合方法;GS融合效果仅次于PCA;而NNDiffuse Pan Sharpening、Wavelet变换和Brovey变换均不适合浮筏识别与提取。
Image fusion is one of the most important steps in remote sensing information extraction.To select the appropriate fusion method is the crucial link.In this paper,Xiangshan Port in Zhejiang Province is the study area,and the oyster culture is the observation target.The satellite of WorldView-2 multispectral and panchromatic images were used to detect the distribution of the coastal oyster farming. The different five fusion methods, such as Principal Component Analysis (PCA),Gram-Schmidt(GS), NNDiffuse Pan Sharpening,Brovey Transform and Wavelet Transform, were evaluated by two of subjective qualitative and objective quantitative aspects.We compared the fused images with the original image using six kinds of statistical parameters including mean, standard deviation, entropy, average grads, correlation coefficient and spectral distortion to evaluate the images' fusion performance.The results indicate that, for the characteristics of coastal oyster farming, the fusion image by principal component analysis method not only preserves detail spatial texture information but also maintains the spectral character well.The method of PCA is the most suitable fusion method for remote sensing appl images.The fusion effect of GS is second to PCA,whi plications.NNDiffuse Pan Sharpening, Wavelet transfo ications in coastal oyster culture with World View-2 ch can be used as an alternate method for fusion aprm and Brovey transform are inappropriate for the identification and extraction of oyster culture floating raft.
作者
周为峰
曹利
李小恕
程田飞
Zhou Weifeng1,2 ,Cao Li1,2 ,Li Xiaoshu3 ,Cheng Tianfei1(1.Key Laboratory of Fishery Resources Remote Sensing and Information Technology ,Chinese Academy of Fishery Sciences, Shanghai 200090, China ; 2.School of Mathematics, Physics and Information Science ,Zhejiang Ocean University ,Zhoushan 316004,China; 3.Scientific Observing and Experimental Station of Fishery Remote Sensing ,Ministry of Agriculture ,Beijing 100041 ,China)
出处
《遥感技术与应用》
CSCD
北大核心
2018年第1期103-109,共7页
Remote Sensing Technology and Application
基金
农业部渔业遥感科学观测实验站开放课题(OFSOESFRS201505)
国家自然科学基金项目(31602206)
上海市自然科学基金项目(16ZR1444700)
中国水产科学研究院东海水产研究所中央级公益性科研院所基本科研业务费项目(2016T05)