Using remote sensing to map in-field variability: Clustering Approaches

Abstract

Precision Farming concept is spreading rapidly worldwide as a tool to enable rmers r protable production while llling environmental and od saty conditions. The introduction and application of precision technologies in agriculture has been motivated by the high degree of variability of agro-ecological conditions within elds. With increased use of precision agriculture techniques inrmation concerning within eld soil and crop variability is becoming increasingly important r eective crop management. As a result of new technological advances, more and more data is available. The sa and reliable transr of dierent kinds of data into detailed inrmation r management purposes is also of increasing importance. ln this paper an overview is given on the use of remote sensing data in site-specic rming r assessing crop growth and yield variability. Many studies have been carried out to nd an appropriate method to classify the high resolution remote sensing data within a eld. According to the results of my research two classication approach - pixel-based and object-based (OBIA) - are presented. Image processing techniques including vegetation indices, segmentation, and classication were used in this research. The models presented in the article are suggested to be used for crop monitoring and supporting decision making.