India has been making waves in the world of information technology.  It is still a long way to go to become an IT superpower.  Recently IT has opened up newer vistas in the field of agricultural research and extension system.  Information is equally vital for research managers, scientists and farmers in order to achieve maximum efficiency aimed at sustained growth of agriculture.  Electronic connectivity has provided  one more channel for mass communication which can be harnessed for extension of agricultural research to reach farmers through mass distance education.

Looking at this vast potential of Information Technology (IT) in 1996, the Saraswati Sugar Mills launched the pilot project CAMREST (Cane Management Through Remote Sensing Technology).  CAMREST project of Saraswati Sugar Mills has major role to play in availing the opportunities created by IT for the benefit of our farmers to develop good quality of sugar cane for the sugar mills.  One of the aims of CAMREST is to develop an accurate, reliable and timely method for generating a agricultural statistics by means of satellite data.  Initially, CAMREST Project was involved to assess the sugarcane acreage, health and production estimation by analysing the satellite images.

Saraswati Sugar Mills, has galvanized the information related technologies such as Satellite Remote Sensing, Geographical Information System (GIS), Global Positioning System (GPS) and Canopy Analyzer, GIS technology offers tremendous scope for informed and planned agricultural decision making GPS identifies accurately various field locations subsequent crop treatment to ascertain healthy crop.  Plant canopy analyzer is a useful tool for yield model development of sugarcane at variety wise on the basis of Leaf Area Index (LAI).

Five satellite images of IRS IC\ID LISS III were taken every year on diferent dates on the basis of sugar cane growth stages and analyzed using the computer aided-digital classification procedure.  The difference in growth stages of sugar cane on given date of data acquisition gives different spectral signatures and was considered as one of the most important means of identifying sugarcane vegetation.  These signatures were used in unsupervised digital classification.  This is turn gave us the sugar cane acerage estimates of 142L, 98K 106L, 114L and 126L acres in year 1196, 1997, 1998, 1999 and 2000 respectively with relative deviation of about 4% as compared to manual survey.  The classification accuracy of training pixels of sugarcane was 94% to 95% and the average accuracy of all the other land use classes was 97%.