IEEE Geoscience and Remote Sensing Magazine - December 2019 - 30

River delta, China, on 1 November 2018. The ecosystem
of the Yellow River delta is characterized by originality,
vulnerability, and rarity because of its significant role in
the connection of atmosphere, rivers, oceans, and land. It
has various species with diverse aquatic, wet, medium, and
salt-tolerant plant communities, e.g., reed, erythrophobia,

Top-of-Atmosphere Radiance

(a)
100

Salt Marsh
Paddy
Architecture
Reed
Spartina
Mudflat

80
60
40
20
0
500

River
Pond
Architecture
Suaeda
Saline

1,000
1,500
2,000
Wavelength (nm)
(b)

2,500

Sea
Salt Marsh
Ditch
Aquafarm
Locust
Paddy
Maize
Reed
Mudflat
Tamarix
Spartina
Soybean
Intertidal Saltwater Marshes
(c)

tamarisk, and spartina alterniflora. Figure 11(b) plots the
spectrum curves of six typical ground objects. The narrow
bands of the AHSI show clear divergences of different objects in the spectral curves. Moreover, Figure 11(c) shows
the classification maps of AHSI hyperspectral images in
the same region. Using top-of-atmosphere radiances as
the input, the support vector machine (SVM) is utilized for
classification. The SVM adopts the radial basis function as
the kernel function, and two parameters of the variance
parameter and penalization factor are estimated via crossvalidation. The 2% of total pixels in each class are randomly selected for training the classifier. The useful classification map supports the previously mentioned observations
and explains the exceptional performance of the AHSI's
hyperspectral imagery for detecting and recognizing different ground objects with subtle spectral divergences.
CONCLUSIONS
This article provided an overview of the AHSI aboard China's GaoFen-5 satellite, including its instrument design, operating principle, and test results, as well as its data-processing chain and sample application. As one six main payloads
on the GaoFen-5 satellite, the AHSI demonstrates China's
advances in spaceborne hyperspectral imaging technology.
It features a large FOV telescope; a low-distortion, large flatfield fine spectrometer; a large infrared focal-plane detector;
a long-life, large cooling-capacity cryocooler; a high-precision calibration system; and a high-precision image-compensation mechanism. It will greatly benefit precise applications such as ecological environment monitoring, natural
resources survey, and oil and gas exploration.
ACKNOWLEDGMENTS
Developmental work on the Advanced Hyperspectral
Imager (AHSI) was performed at the Shanghai Institute
of Technical Physics and the Chinese Academy of Sciences and was supported by the Major Project of National
High-Resolution Earth Observation System of China under grant A0106/1112. The authors would like to thank
the collaborative teams from the Satellite Environment
Center; the Ministry of Ecology and Environment; China
Aero Geophysical Survey and Remote Sensing Center for
Natural Resources; the State Administration of Science,
Technology and Industry for National Defense; the China Center for Resources Satellite Data and Application;
China Aerospace Science and Technolog y Corporation; the Shanghai Academy of Spaceflight Technology;
and the Shanghai Institute of Satellite Engineering for
their valuable suggestions and sincere help in developing the AHSI.

FIGURE 11. The sampled AHSI hyperspectral data in the Yel-

low River delta of China. (a) The VNIR true-color red-green-blue
composite image (R: 630 nm, G: 557 nm, and B: 497 nm), (b) the
sample spectra for typical targets, and (c) the classifications of
different ground features. (Source: obtained by AHSI; used
with permission.)
30

AUTHOR INFORMATION
Yin-Nian Liu (ynliu@mail.sitp.ac.cn) received his Ph.D.

degree in remote sensing from the Shanghai Institute of
Technical Physics (SITP), Chinese Academy of Sciences (CAS), in 2005. He is the primary investigator of the
IEEE GEOSCIENCE AND REMOTE SENSING MAGAZINE

DECEMBER 2019

IEEE Geoscience and Remote Sensing Magazine - December 2019

Table of Contents for the Digital Edition of IEEE Geoscience and Remote Sensing Magazine - December 2019