Contact  |  Chinese  |  CAS
 Home  About RADI People Research News Cooperation Education Resources Data Services
Location:Home > Research > Research Progress
China’s Intelligent Hyperspectral Remote Sensing Research Attracts World Attention
 
 Date: 2015-08-03  Page Views:
Print
Close

At the IEEE 7th Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing (IEEE WHISPERS 2015) held recently in Tokyo, Japan, RADI researcher ZHANG Bing was invited to give a plenary speech on “Intelligent Hyperspectral Remote Sensing Satellite (IHRS): A New Perspective”. Prof. Zhang shared the research results in the optimization of intelligent hyperspectral remote sensing imaging mode, intelligent analysis of images and real-time data processing and their huge potentials in future application.

The hyperspectral remote sensing research started in the 1980s and its results have been widely applied to precise identification of ground objects and qualitative parameter inversion. But traditional hyperspectral remote sensing, though capable of providing abundant spectral images of ground objects, has difficulty in distributing and processing massive data and its information extraction model is too complex. All these factors have to certain extent restricted its application.

Prof. Zhang and his team propose the concept of intelligent hyperspectral remote sensing satellite system and establish the conversion model for the optimized ground elements observation mode. The results of the test on six typical altered minerals including kaolinite show that the usage of intelligent observation could increase the diction rate from 68.5% to 80.9%.  The project team has established the intelligent, optimized mixed pixel group decomposition model and solved problems with ground information extraction at the sub-pixel level and optimal classification. In the experiment, they reduced the endmember spectral angle error from 5.636°to 5.023°, and the inversion error from 4.267 to 2.586, and explored a new direction for the mixed pixel decomposition research. They've also proposed methods such as Adjusted Spectral Matched Filter (ASMF) which have remarkably improved the precision of target detection.

In addition, they've solved many difficulties with real-time processing of on-planet hyperspectral images, developed China’s first FPGA-based on-planet real-time hyperspectral image processing system, and for the first time synchronized data acquiring with data processing.

The above research results have been published on internationally prestigious journals on remote sensing including IEEE TGRS, IEEE JSTARS, IEEE GRS Letters, Remote Sensing, and JARS since 2011, producing 20 SCI papers and eight national invention patents.

For the introduction to the plenary report, please visit:

http://www.ieee-whispers.com/index.php/infos/plenary-speakers/21-plenary-speaker-3

Downloads:
[ Top ]  
Copyright © 2013 by the Institute of Remote Sensing and Digital Earth
No.9 Dengzhuang South Road, Haidian District, Beijing 100094, China Tel:86-10-82178008 Fax:86-10-82178009