Title :
An Integrated System for Mapping Red Clover Ground Cover Using Unmanned Aerial Vehicles: A Case Study in Precision Agriculture
Author :
Abuleil, Ammar M. ; Taylor, Graham W. ; Moussa, Medhat
Author_Institution :
Sch. of Eng., Univ. of Guelph, Guelph, ON, Canada
Abstract :
In the field of precision agriculture (PA), Un-manned Aerial Vehicles (UAVs) are creating new opportunities for remotely assessing various characteristics of crops. In this paper, we present two main contributions that were evaluated on a novel application: mapping red clover ground cover (RCGC). First, we develop an integrated system for collecting, pre-processing and analyzing aerial data for the mapping of RCGC at a patch-level. Second, we collected, ground-trusted, and pre-processed a RCGC dataset that we make public for further analysis. We evaluated several different machine learning classifiers for mapping image patches to discrete clover coverage levels, reaching an accuracy of 91%.
Keywords :
agriculture; autonomous aerial vehicles; image classification; learning (artificial intelligence); path planning; precision engineering; robot vision; PA; RCGC mapping; UAV; aerial data analysis; aerial data collection; aerial data preprocessing; crop characteristics; image patch mapping; machine learning classifiers; precision agriculture; red clover ground cover mapping; unmanned aerial vehicles; Accuracy; Agriculture; Data collection; Global Positioning System; Hyperspectral sensors; Sensors; Support vector machines; classification; ground cover; machine learning; precision agriculture; red clover; remote sensing;
Conference_Titel :
Computer and Robot Vision (CRV), 2015 12th Conference on
Conference_Location :
Halifax, NS
DOI :
10.1109/CRV.2015.43
