Share Email Print
cover

Proceedings Paper

Early pest detection in soy plantations from hyperspectral measurements: a case study for caterpillar detection
Author(s): Matías Tailanián; Enrique Castiglioni; Pablo Musé; Germán Fernández Flores; Gabriel Lema; Pedro Mastrángelo; Mónica Almansa; Ignacio Fernández Liñares; Germán Fernández Liñares
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

Soybean producers suffer from caterpillar damage in many areas of the world. Estimated average economic losses are annually 500 million USD in Brazil, Argentina, Paraguay and Uruguay. Designing efficient pest control management using selective and targeted pesticide applications is extremely important both from economic and environmental perspectives. With that in mind, we conducted a research program during the 2013-2014 and 2014-2015 planting seasons in a 4,000 ha soybean farm, seeking to achieve early pest detection. Nowadays pest presence is evaluated using manual, labor-intensive counting methods based on sampling strategies which are time consuming and imprecise. The experiment was conducted as follows. Using manual counting methods as ground-truth, a spectrometer capturing reflectance from 400 to 1100 nm was used to measure the reflectance of soy plants.

A first conclusion, resulting from measuring the spectral response at leaves level, showed that stress was a property of plants since different leaves with different levels of damage yielded the same spectral response. Then, to assess the applicability of unsupervised classification of plants as healthy, biotic-stressed or abiotic-stressed, feature extraction and selection from leaves spectral signatures, combined with a Supported Vector Machine classifier was designed. Optimization of SVM parameters using grid search with cross-validation, along with classification evaluation by ten-folds cross-validation showed a correct classification rate of 95%, consistently on both seasons. Controlled experiments using cages with different numbers of caterpillars--including caterpillar-free plants--were also conducted to evaluate consistency in trends of the spectral response as well as the extracted features.

Paper Details

Date Published: 14 October 2015
PDF: 9 pages
Proc. SPIE 9637, Remote Sensing for Agriculture, Ecosystems, and Hydrology XVII, 96372I (14 October 2015); doi: 10.1117/12.2195083
Show Author Affiliations
Matías Tailanián, CSI Ingenieros (Uruguay)
Enrique Castiglioni, Univ. de la República (Uruguay)
Pablo Musé, CSI Ingenieros (Uruguay)
Univ. de la República (Uruguay)
Germán Fernández Flores, CSI Ingenieros (Uruguay)
Gabriel Lema, CSI Ingenieros (Uruguay)
Pedro Mastrángelo, CSI Ingenieros (Uruguay)
Mónica Almansa, CSI Ingenieros (Uruguay)
Ignacio Fernández Liñares, CSI Ingenieros (Uruguay)
Germán Fernández Liñares, CSI Ingenieros (Uruguay)


Published in SPIE Proceedings Vol. 9637:
Remote Sensing for Agriculture, Ecosystems, and Hydrology XVII
Christopher M. U. Neale; Antonino Maltese, Editor(s)

© SPIE. Terms of Use
Back to Top