- You are here:
-
-
All publications
- Preparing for machine learning model to predict sufficiency of Boron in potato leaves using field-measured spectral data
All publications from your selection
Preparing for machine learning model to predict sufficiency of Boron in potato leaves using field-measured spectral data
Authors: Sama Huseynova,
Ahmad Al Mallahi
Identifier: CSBE23158
Download file:
Published in: CSBE-SCGAB Technical Conferences » AGM Lethbridge 2023
Download RAW file:
DOI:
Description: Plant growth and development mainly depend on the soil's mineral nutrient composition and concentration. Due to their immobility, plants may not obtain adequate supply of specific nutrients. As such, foliar application of certain nutrients such as boron is done during the growing season to compensate for any possible deficiency and enhance potato production. Currently, analyzing plant tissue is done at external laboratories based on petiole chemical testing whose results take nearly two weeks to reach the grower, hence the decision to spray is based solely on the grower?s experience. Therefore, this research aims to develop a machine learning model to determine the levels of sufficiency and deficiency of boron based on its contents in petioles and relating them to leaf reflectance. We collected samples from three locations in New Brunswick to create a dataset of 98data points, each of which, consists of 50-70 petioles and their tip leaves. The leaves were scanned using a Vis-NIR spectrophotometer that covers the range of 400-2500 nm, whereas the chemical analysis is done at the laboratory, whose guidelines divide the boron status into 3 categories, excessive, sufficient, and deficient. The datasets for machine learning were arranged based on this categorization. Next, neural networking classification will be conducted to find out the possibility of sensing the boron status using spectral reflectance. The ability to get this information using a sensor will provide the nutrient status in a timely manner so that foliar application of boron is decided on the basis of data instead of guessing.
Keywords: nutrient deficiency, boron, machine learning, classification
Résumé:
Mots-clés:
Citation:
Volume:
Issue:
Pages -
Contributor:
Date: 2023-07-23
Technical field:
Conference name: CSBE/SCGAB 2023 Annual Conference, Lethbridge, Alberta, 23-26 July 2023.
Session name:
Other information:
Type: Presentation
Format:
Publication type: Text.Abstract
Source:
Relation:
Coverage: North_America
Language 1: en
Language 2:
Rights: Canadian Society for Bioengineering
Notes:
Other files:
DOI:
Description: Plant growth and development mainly depend on the soil's mineral nutrient composition and concentration. Due to their immobility, plants may not obtain adequate supply of specific nutrients. As such, foliar application of certain nutrients such as boron is done during the growing season to compensate for any possible deficiency and enhance potato production. Currently, analyzing plant tissue is done at external laboratories based on petiole chemical testing whose results take nearly two weeks to reach the grower, hence the decision to spray is based solely on the grower?s experience. Therefore, this research aims to develop a machine learning model to determine the levels of sufficiency and deficiency of boron based on its contents in petioles and relating them to leaf reflectance. We collected samples from three locations in New Brunswick to create a dataset of 98data points, each of which, consists of 50-70 petioles and their tip leaves. The leaves were scanned using a Vis-NIR spectrophotometer that covers the range of 400-2500 nm, whereas the chemical analysis is done at the laboratory, whose guidelines divide the boron status into 3 categories, excessive, sufficient, and deficient. The datasets for machine learning were arranged based on this categorization. Next, neural networking classification will be conducted to find out the possibility of sensing the boron status using spectral reflectance. The ability to get this information using a sensor will provide the nutrient status in a timely manner so that foliar application of boron is decided on the basis of data instead of guessing.
Keywords: nutrient deficiency, boron, machine learning, classification
Résumé:
Mots-clés:
Citation:
Volume:
Issue:
Pages -
Contributor:
Date: 2023-07-23
Technical field:
Conference name: CSBE/SCGAB 2023 Annual Conference, Lethbridge, Alberta, 23-26 July 2023.
Session name:
Other information:
Type: Presentation
Format:
Publication type: Text.Abstract
Source:
Relation:
Coverage: North_America
Language 1: en
Language 2:
Rights: Canadian Society for Bioengineering
Notes:
Other files:
©
2026
CSBE-SCGAB