One method for HJ-1-A HSI and CCD data fusion

35th International Symposium on Remote Sensing of Environment

地球科学;生态环境科学

Xiong, Wencheng^1,2 ; Shao, Yun^1 ; Shen, Wenming^2 ; Xiao, Rulin^2 ; Fu, Zhuo^2 ; Shi, Yuanli^2

Institute of Remote Sensing Application, Chinese Academy of Sciences, Beijing, China^1

Satellite Environment Application Center, Ministry of Environment Protection, Beijing, China^2

关键词: Classification tests;    Detection algorithm;    High spatial resolution;    High spectral resolution images;    Multispectral images;    Multispectral sensors;    Spectral classification;    Spectral information;   

Others  :  https://iopscience.iop.org/article/10.1088/1755-1315/17/1/012226/pdf
DOI  :  10.1088/1755-1315/17/1/012226

学科分类：环境科学（综合）

来源: IOP

【 摘 要 】

HJ-1-A satellite, developed by China independently, was equipped with two sensors of Hyper Spectral Imager (HSI) and multispectral sensor (CCD). In this paper, we examine the benefits of combining data from CCD data (high-spatial-resolution, low-spectral-resolution image) with HSI data (low -spatial-resolution, high -spectral-resolution image). Due to the same imaging time and similar spectral regime, the CCD and HSI data can be registered with each other well, and the difference between CCD and HSI data mainly is systematic bias. The approach we have been investigating compares the spectral information present in the multispectral image to the spectral content in the hyperspectral image, and derives a set of equations to approximately acquire the systematic bias between the two sensors. The systematic bias is then applied to the interpolated high-spectral CCD image to produce a fused product. This fused image has the spectral resolution of the hyperspectral image (HSI) and the spatial resolution of the multispectral image (CCD). It is capable of full exploitation as a hyperspectral image. We evaluate this technique using the data of Honghe wetland and show both good spectral and visual fidelity. An analysis of SAM classification test case shows good result when compared to original image. All in all, the approach we developed here provides a means for fusing data from HJ-1-A satellite to produce a spatial-resolution-enhanced hyperspectral data cube that can be further analyzed by spectral classification and detection algorithms.