APPLICATION OF THE MEMORY-PREDICTION THEORY IN OLIVE GROVE INVENTORY

Abstract

Remote sensing using aerial or satellite images represents an interesting alternative to the traditional manual procedures used for agricultural inventory by incorporating automatic processes. This paper presents an inference system for olive grove detection using aerial photographs. The system is inspired by a recent memory-prediction theory and models of high-level architecture of the human neocortex. This paper describes the hierarchical architecture and recognition performance of this Bayesian model. Results indicate that 95% of the olive groves are detected by using images obtained from a photogrammetric sensor. It is concluded that the automatic process developed can be easily integrated into the final user's Geographical Information System and produces useful information for olive grove management.