Automatic identification of the area covered by acorn trees in the dehesa (pastureland) Extremadura of Spain

Abstract

The acorn is the fruit of oak trees and is an important crop in the Spanish dehesa extremeña because of the value it provides to feeding Iberian pigs to obtain the “acorn-fed” certification. For this reason, it is desirable to maximise production of Iberian pigs with the appropriate weight. Knowing the area covered by the acorn treetops to determine the Tree-Covered Area (TCA), a value required to estimate the number of Iberian pigs that can be released per hectare, as regulated by the Royal Decree 4/2014, issued by Spain in January of 2014. In this work, we propose a method to automatically estimate the TCA through digital aerial images (orthophotos) of the pastureland of Extremadura and, with this, to facilitate determining the number of Iberian pigs that can be released in a specific parcel of land. Requirements for automatic detection include the correct identification of acorn trees, correctly discerning their shadows, and discerning arbuscles (young acorn trees that are still unproductive, or shrubs that are not oaks). These requirements represent challenges that apply to both automatic and manual segmentations. The automated segmentation in this study is based on the modified version of the Gustafson-Kessel clustering algorithm presented by Babuska. The results of this work are promising when compared with the actual images and when compared with the results of manual segmentation. The entire set of 38 aerial images used in this work correspond to an approximate area of 15.56 hectares, and the results are of great interest to producers of Iberian pork with the “acorn-fed” certification.