International Journal of Science and Research (IJSR) ISSN (Online): 2319-7064 Index Copernicus Value (2015): 78.96 | Impact Factor (2015): 6.391 Volume 6 Issue 6, June 2017 www.ijsr.net Licensed Under Creative Commons Attribution CC BY Study of Bangladesh Soil Nutrient over Satellite Images Sabbir M Saleh 1 , S M Raisul Islam 2 , Md. Imran Hossain 3 , Khondoker Ali Asgor Pavel 4 1 University of South Asia, Lecturer, Department of Computer Science and Engineering, Banani, Dhaka – 1216, Bangladesh 2 University of Dhaka, Graduated, Management Information System, Dhaka, Bangladesh 3 American International University- Bangladesh, Graduated, Electrical & Electronic Engineering, Banani, Dhaka – 1216, Bangladesh 4 University of Dhaka, Graduated, Institute of Information Technology, Dhaka, Bangladesh Abstract: In the Asian, African and Latin American countries, well over 500 million [1] people are living in what the World Bank has called "absolute poverty". Nearly one in four people, 1.3 billion - a majority of humanity - live on less than $1 per day [2] – not even enough to feed themselves well. Suffering from hunger many of them die gradually. A solution to this problem could be letting them able to produce enough amount of foods. Using satellite imagery, it is possible to analyze the amount of nutrient available in soil. If soil properties is known, it’s possible to determine which crop will grow best and what amount of fertilizer is needed. Obviously it will reduce the cost and maximize the food production. Keywords: Remote sensing, Hyper spectral, ETM+, Digital Number (DN), Radiance, Reflectance, Soil Nutrient. 1. Introduction The whole world has been suffering from many problems, most probably, from the first day of its formation. Among all, one of the major problems is hunger. Researchers, scholars, scientists have been trying to solve this problem from many days. And significant goals have also been achieved in this sector. As the time passed on, the solutions have become more efficient and productive. However, hunger is still one of the major problems because of the continuous growth of population and also the natural or human made disasters. It is not humanly possible to reduce this massive growth of population, or predict and avoid the disasters, or increase the size of the cultivating land. So the only way out of this is to use the available resources and make them more productive. The conventional agricultural system in Bangladesh and also most of the world does not consider the spatial variability of soil and crops. In spite of that, fertilizers, pesticides and irrigation are applied uniformly on the field. As a result, there is a greater possibility that, the field would suffer from excessive usage of chemicals, which is harmful to the crops. To increase production, one needs to know all the factors that can affect the growth of crops in both good and bad way. For example, the type of soil, the amount of soil nutrients, the height of the cultivating land, the season, flood, drought and many more factors influence it severely. And also not to forget that, the amount of soil nutrients present in a region can change in a certain period of time. It is a must to have an accurate knowledge on the region of cultivation. However, it is very hard to cope-up with these variations, because the traditional procedure of soil sampling that is used in most of the countries including Bangladesh is time consuming, labor intensive and expensive. Remote sensing process can In the remote sensing process, the data captured by remote sensing systems is analyzed by interpretive and measurement techniques in order to provide useful information. These techniques are diverse, ranging from traditional methods of visual interpretation to methods using sophisticated computer processing. Remote sensing is a technique of obtaining information about objects by analyzing data - collected using tools that are not physically in contact with the objects. An example of remote sensing device is satellite. Satellites capture images of the surface while rotating in its orbit. The images captured by the satellites consist of several monochrome images of the same object. As example, LANDSAT5 [3] produces 7 band images with the wavelength of the bands being between 450nm and 1250nm [4]. LANDSAT7 [5] generates 8band images. This paper exhibits an analysis on the variability of soil throughout a region and also an argument - if it is possible to determine the amount of specific soil nutrients that is present in that region. It is focused on examining the soil using hyper spectral reflectance data and predicting the soil nutrients. 2. Literature Review Remote sensing technique is being used widely in agriculture filed. An example could be precision farming [9]. Many researchers researched on determining factors of cultivation using remote sensing and satellite imagery. J.G.P.W. Clevers, L. Kooistra and M.E. Schaepman showed in their research [6] that it is possible to detect canopy water using remote sensing. They used PROSAIL to measure canopy water. Fouad Al-Khaier [7] described and compared two methods of determining soil salinity using remote sensing. Fouad Al- Khaier, in his research, used biophysical [7] method for Landsat 7 image. Biophysical method is mainly built on the concept of surface resistance. He showed the relation between surface resistance and soil salinity. Paper ID: ART20173196 DOI: 10.21275/ART20173196 1021