Remote Sensing Satellite Data Analysis applied to precision Agriculture Modeling

Abstract

This research paper focuses on analysis of Image of Data from Remote Sensing which integrate Spatial, Temporal and Spectral Features of the objects in the field of Satellite image processing. We used remote sensed data in multi spectral form to detect spectral signature related to several objects in several regions used in land cover classification, change in land use with respect to time and using temporal analysis for analysis the effect of climate upon the surface. Few band combinations of data from remote sensing are very useful in classification for land cover. Land cover types having spatial distributions like; agriculture land, urban area and water resource are easy to interpret.Method used to derive agriculture variables from different data through remote sensing to extract quantitative parameters using several multispectral sensors in remote sensing, issues related to inter-calibration in different methods are to be considered for the assurance of comparability. Some features affects vegetation indices like; atmospheric conditions, sensor geometry, radiometric or spatial resolution and topography.Factors involved in this research are spectral characteristics related to different sensors viz- bandwidth, centre wavelengths, and band position, and may be described by relative spectral response functions. Vegetation Indes (VI) and Weighted Difference Vegetation Indes (WDVI) in between several sensor pairs using regression are based on multispectral sensors in simulated form. The method which combines the information in multi-spectral form with calculation was not able to provide good results and was beaten by use of single sensor, in the case of multispectral information. The results of remote sensing and statistical data are used for estimation of input parameters for application in agriculture. Input parameter for agriculture extracted from the data of remote sensing is main advantage in this technique as a large area of spatial overview is taken.Next challenge was for integration of the agriculture variables extracted from data in multispectral form into model for agriculture growth for increase the accuracy of estimation for final yield. At present the linking for our benefit between these two techniques is limited for classification of land use using remote sensing to adopt the appropriate model and agriculture growth quantification as well as development curves with the use of biophysical parameters extracted from images from remote sensing for calibration of models. Hence variables of remote sensing are integrated with agriculture growth are modeled for improving the accuracy of estimation of final yield as compared to parameter setting as automatic input for obtaining optimized yield.Hence data of remote sensing integration with model of agriculture growth facilitates the spatial application in prediction at best scale for agriculture production. Above approach is best one with any other evaluated method of yield estimation using direct multi sensors. This research has proved the retrieval of biophysical parameters from the data of remote sensing and used to prepare model for agriculture growth, for improved estimation of final yield for sustainable agricultural production for developed as well as developing countries.