Hyperspectral Data Classification using Geostatistics and Support Vector Machines

Remote Sensing Letters, Volume 2, Issue 2 2011 , pages 99 – 106

S. Bahria; N. Essoussi; M. Limam

“Hyperspectral imagery combined with spatial features holds promise for improved remote sensing classification. In this letter, we propose a method for classification of hyperspectral data based on the incorporation of spatial arrangement of pixel’s values. We use the semivariogram to measure the spatial correlation which is then combined with spectral features within the stacked kernel support vector machine framework. The proposed method is compared with a classifier based on first-order statistics. The overall classification accuracy is tested for the AVIRIS Indian Pines benchmark dataset. Error matrices are used to estimate individual class accuracy. Statistical significance of the accuracy estimates is assessed based on the kappa coefficient and z-statistics at the 95% confidence level. Empirical results show that the proposed approach gives better performance than the method based on first-order statistics.”

Jack Dangermond to Receive Global Citizen Award at GSDI 12

Jack Dangermond, founder and President of Esri, has been named to receive the Global Citizen Award of the GSDI Association. The award will be presented on the opening day of the GSDI 12 World Conference in Singapore (http://gsdi.org/gsdi12) where Jack, as the recipient, will give the lead keynote address. His presentation on October 19 will focus on a global vision for spatially enabling society.

The Global Citizen Award is an occasional award of the GSDI Association and recognizes an individual who has provided exemplary thought leadership and substantive worldwide contributions in (1) promoting informed and responsible use of geographic information and geospatial technologies for the benefit of society and (2) fostering spatial data infrastructure (SDI) developments or geospatial advancements supporting sustainable social, economic, and environmental systems integrated from local to global scales. A recipient should be an exemplary citizen in substantively serving the needs of others internationally or globally, contribute selflessly to his or her profession, and contribute as well to his or her own community, state and nation. Jack’s specific contributions will be highlighted at the award ceremony.

The Global Spatial Data Infrastructure (GSDI) Association is an inclusive organization with members from all sectors – public and private – and individuals from around the world promoting international cooperation and collaboration in support of local, national, regional, and international spatial data infrastructure developments.  The goal of the GSDI Association is to allow nations to better address social, economic, and environmental issues of pressing importance through the establishment and use of SDIs and geospatial information. Please see http://gsdi.org for more information.

[Source: GSDI News]

Influence of Topography on the Endemicity of Kala-azar: A Study based on Remote Sensing and Geographical Information System

Geospatial Health, Volume 4, Number 2, May 2010, Pages 155-165

Gouri S. Bhunia,  Shreekant Kesari,  Algarsamy Jeyaram,  Vijay Kumar,  Pradeep Das

“Kala-azar, a fatal infectious disease in many Indian states, particularly in Bihar, West Bengal, Uttar Pradesh, and Jharkhand, is caused by the protozoan parasite Leishmania donovani and transmitted by the sandfly vector Phlebotomus argentipes. The vector is distributed all over the country but the disease is confined to particular zones since before the last century. In this study, parameters such as altitude, temperature, humidity, rainfall and the normalized difference vegetation index (NDVI) were investigated for correlation with the distribution of the disease in the northeastern corner of the Indian sub-continent. Data analysis on Kala-azar prevalence during the period 2005-2007 in the four states showed that the highest prevalence was below 150 m of altitude with very few cases located above the 300 m level. Low NDVI value ranges (0.03-0.015) correlated with a high occurrence of the disease. The maximum temperatures in the affected sites varied between an upper level of 25-29°C and a minimum of 16-20°C. The rainfall in these areas fluctuated between 1154 and 1834 mm. As the disease showed a high correlation with the prevailing topographic conditions, an attempt was made to improve the relative strength of the approach to predict the potential for endemicity of leishmaniasis by introducing satellite imagery complemented with a geographical information system database.”

The Sensor Web: Systems of Sensor Systems

International Journal of Digital Earth, Volume 2, Issue 1 March 2009, pages 16 – 30

T. L. van Zyl; I. Simonis; G. McFerren

“Global Earth Observing System of Systems (GEOSS) presents a great challenge of System of Systems integration across organisational and political boundaries. One existing paradigm that can address the scale of the challenge is that of the Sensor Web. In this paradigm, the internet is evolving into an active, macro sensing instrument, capable of drawing sensory data from around the globe to the fingertips of individuals. The Sensor Web will support scientific research and facilitate transparent political decision making. This article presents some of the technologies explored and activities engaged in by the GEOSS Sensor Web community, towards achieving GEOSS goals.”