Hyperspectral Input-Based Neural Network Model for Predicting Chlorophyll Status in Corn

Siza D. Tumbo, David G. Wagner, Paul Heinz Heinemann

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Scopus citations

Abstract

The use of spectral reflectance techniques for predicting nitrogen in corn at V6 growth stage has been limited by the soil background, and changes in cloud cover and solar angles. Measurement and prediction techniques, which are independent of these factors, are needed for fast and accurate prediction of nitrogen deficiency at V6 growth stage for site-specific side dressing of nitrogen. Spectral reflectance response patterns (SRRPs) from individual corn plants were collected under variable cloud cover and solar angles using a fiber optic spectrometer. Chlorophyll levels, which are strong indicators of nitrogen status in plants, were also measured on each corn plant using a SPAD chlorophyll meter. The back-propagation neural network model was trained using spectral channels of the SRRPs as inputs and chlorophyll readings as outputs. The model showed strong correlation between predicted and actual chlorophyll meter readings (r 2=0.91, root mean square prediction error = 1.30 in SPAD units with validation set).

Original languageEnglish (US)
Title of host publication2000 ASAE Annual International Meeting, Technical Papers: Engineering Solutions for a New Century
Pages1627-1642
Number of pages16
Volume1
StatePublished - 2000
Event2000 ASAE Annual International Meeting, Technical Papers: Engineering Solutions for a New Century - Milwaukee, WI., United States
Duration: Jul 9 2000Jul 12 2000

Other

Other2000 ASAE Annual International Meeting, Technical Papers: Engineering Solutions for a New Century
CountryUnited States
CityMilwaukee, WI.
Period7/9/007/12/00

All Science Journal Classification (ASJC) codes

  • Engineering(all)

Fingerprint Dive into the research topics of 'Hyperspectral Input-Based Neural Network Model for Predicting Chlorophyll Status in Corn'. Together they form a unique fingerprint.

Cite this