Definition of Management Zones Through Image Processing for Precision Agriculture

Abstract

The continuous growth of the world population results in ever growing demands on food production, however the amount of area available to expand farming activities gets smaller every day. This is the chief motivator for techniques to help farmers improve the efficiency of their crops. This paper proposes the utilization of aerial images of plantations, captured by unmanned aerial vehicles (UAV), in order to monitor the plant’s health, the vitality of the vegetation is calculated using well-known methods such as Normalized Difference Vegetation Index (NDVI). However, alone this data is not enough because the farmer cannot decide which actions and where he should take, for that reason the management zones are needed. The vigor zones are calculated utilizing a classification algorithm to divide the crop in a prefixed number of regions, grouping those with similar aspects, which allows the producer to decide how he will act in each zone.