Modeling zone management in precision agriculture through Fuzzy C-Means technique at spatial database

Date
2011-08-01
Authors
Vendrusculo, Laurimar
Kaleita, Amy
Kaleita, Amy
Major Professor
Advisor
Committee Member
Journal Title
Journal ISSN
Volume Title
Publisher
Altmetrics
Authors
Research Projects
Organizational Units
Journal Issue
Series
Department
Agricultural and Biosystems Engineering
Abstract

Predict the optimal number of zones to manage tasks evolved in precision agriculture applications is challenging issue in classification tasks. Important decisions in the farm required maps of yield classes which contain relative large, similar and spatially contiguous partitions and sometimes without a priori knowledge of the field. The main goal of this study was to apply Fuzzy C-means (FCM), an unsupervised classification technique, in a geo-referenced yield and grain moisture dataset in order to find optimal number for homogeneous zones. Those data were produced by Long-Term Ecological Research in a Biological Station (KBS-LTER), Michigan, during growing season at 2008. The best results presented by this algorithm ranged from 8 to 10 zones which were validated using the indexes Partition Coefficient (PC), Classification Entropy (CE) and Dunn’s Index (DI). Even though, only two attributes were collected in the dataset, the Fuzzy C-means has shown promissing results for zone mapping.

Comments

This is an ASABE Meeting Presentation, Paper No. 1111774.

Description
Keywords
Citation
DOI
Source