May 9, 2009
Center for Computational Engineering and Networking, Coimbatore
Research associates from CEN participated in the 10th ESRI India User Conference conducted on April 28-29, 2009 at Noida in New Delhi. ESRI is a US-based company providing GIS (Geographic Information Systems) software and applications. They have hosted user conferences in various countries during the past decade.
CEN papers were accepted in the conference track – Remote Sensing and Wireless Networks. Robust Watermarking of Remote Sensing Images without Loss of Spatial Information — paper by T. Hemalatha, V. Joevivek, K. Sukumar, K. P. Soman was awarded the second-best paper presentation in the conference.
Details of other papers presented at the conference are given below -
P. Geetha, V. Sowmya, V. Vipin, K. P. Soman, An Approach to Study the Influence of Fish Abundance in Relation to Temperature using GIS – A Case Study of Aliyar Reservoir, Tamil Nadu.
K. Sukumar, V. Joevivek, T. Hemalatha, K. P. Soman, Prediction of Forest Fires using SVM – A Novel Approach.
In environmental studies and earth observations, remote sensing images play a vital role. Spatial and spectral information present in these images are useful for many applications. It is important to protect this information while watermarking a remote sensing image.
There are many robust watermarking algorithms in use today that embed the watermark imperceptibly in the image. However, they do not take into account how the spatial and spectral information may be changed. In this paper, a wavelet-based watermarking algorithm was proposed. This would preserve spatial and spectral information of remote sensing images.
Quality of the watermarked image is determined by performing data comparison operations in ArcGIS. Proposed method is checked for robustness by performing attacks STIRMARK benchmark ,  and ArcGIS. It was shown that the proposed method is robust enough to withstand the hostile attacks. A comparative study was also done between proposed method and method ; the former gave a more promising result.
Geographic Information Systems (GIS) is a rapidly growing technological field that incorporates graphical features with tabular data in order to access real world problems. It has gained importance in the area of fisheries science and management. The GIS community has an opportunity to impact positively the world-wide fisheries crisis. Its usefulness in mapping makes it an important tool for fishermen.
The GIS technique using ARCGIS 9.2. was applied to the Aliyar reservoir. A GIS data case comprising fish abundance in relation to temperature was created. The study provided an enormous amount of knowledge about the abundance of fish in relation to the temperature in Aliyar reservoir. Since India has considerable livestock and poultry population, efforts have to be mobilized to reclaim the resources and put them to effective use.
Forest fires are common hazards that pose a threat not only to forest wealth but also to human lives. A recent survey conducted by The Forest Survey of India found that fires annually affect around 55% of forest area in India. Forest fires are a major cause of forest degradation in the country.
The only effective way to minimize damage caused by forest fires is in their early detection and fast reaction. Traditionally this was done by the human surveillance. This paper proposed an SVM-based classification model that can predict forest fires well before they happen by learning process using sensor information and image captures of video cameras sensitive in visible spectra.