An investigation into the potential of Gabor wavelet features for scene classification in wild blueberry fields

A Gabor wavelets based technique was investigated as a potential tool for scene classification (into one of bare patch, plant, or weed) for its ultimate utility in site-specific application of agrochemicals in wild blueberry fields.Images were gathered from five sites located in central Nova Scotia,...

Full description

Bibliographic Details
Main Authors: Gashaw Ayalew, Qamar Uz Zaman, Arnold W. Schumann, David C. Percival, Young Ki Chang
Format: Article
Language:English
Published: KeAi Communications Co., Ltd. 2021-01-01
Series:Artificial Intelligence in Agriculture
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2589721721000155
id doaj-9644381eebe944df9837102f9150f99e
record_format Article
spelling doaj-9644381eebe944df9837102f9150f99e2021-04-18T06:32:27ZengKeAi Communications Co., Ltd.Artificial Intelligence in Agriculture2589-72172021-01-0157281An investigation into the potential of Gabor wavelet features for scene classification in wild blueberry fieldsGashaw Ayalew0Qamar Uz Zaman1Arnold W. Schumann2David C. Percival3Young Ki Chang4Independent Researcher, Thorold, ON L2V 1Z8, Canada; Corresponding author.Engineering Department, Dalhousie University Faculty of Agriculture, Truro, NS B2N 5E3, CanadaCitrus Research and Education Centre, University of Florida, Lake Alfred, FL33850, USAPlant, Animal and Environmental Sciences Department, Dalhousie University Faculty of Agriculture, Truro, NS B2N 5E3, CanadaEngineering Department, Dalhousie University Faculty of Agriculture, Truro, NS B2N 5E3, CanadaA Gabor wavelets based technique was investigated as a potential tool for scene classification (into one of bare patch, plant, or weed) for its ultimate utility in site-specific application of agrochemicals in wild blueberry fields.Images were gathered from five sites located in central Nova Scotia, Canada. Gabor wavelet features extracted from these images were used to classify scenes according to visually determined classes using step-wise linear discriminant analysis.For individual fields, classification accuracy attained ranged between 87.9% and 98.3%; selected Gabor features ranged between 27 and 72; contextual accuracy for herbicide ranged between 67.5% and 96.7%, and contextual accuracy for fertilizer ranged between 63.6% and 97.1%. The pooled scenes yielded a classification accuracy of 81.4%, and contextual accuracy figures of 61.1% and 73.1% for herbicide and fertilizer, respectively, with selected Gabor features of 36.Calibrations based on LDA coefficients from the pooled scenes could help avoid the need to re-calibrate for each field, whereas those based on individual field LDA coefficients could improve accuracy, hence enable saving on expensive agrochemicals.http://www.sciencedirect.com/science/article/pii/S2589721721000155Wavelet transformsWild blueberryComputer visionMachine learningDiscriminant analysisPrecision agriculture
collection DOAJ
language English
format Article
sources DOAJ
author Gashaw Ayalew
Qamar Uz Zaman
Arnold W. Schumann
David C. Percival
Young Ki Chang
spellingShingle Gashaw Ayalew
Qamar Uz Zaman
Arnold W. Schumann
David C. Percival
Young Ki Chang
An investigation into the potential of Gabor wavelet features for scene classification in wild blueberry fields
Artificial Intelligence in Agriculture
Wavelet transforms
Wild blueberry
Computer vision
Machine learning
Discriminant analysis
Precision agriculture
author_facet Gashaw Ayalew
Qamar Uz Zaman
Arnold W. Schumann
David C. Percival
Young Ki Chang
author_sort Gashaw Ayalew
title An investigation into the potential of Gabor wavelet features for scene classification in wild blueberry fields
title_short An investigation into the potential of Gabor wavelet features for scene classification in wild blueberry fields
title_full An investigation into the potential of Gabor wavelet features for scene classification in wild blueberry fields
title_fullStr An investigation into the potential of Gabor wavelet features for scene classification in wild blueberry fields
title_full_unstemmed An investigation into the potential of Gabor wavelet features for scene classification in wild blueberry fields
title_sort investigation into the potential of gabor wavelet features for scene classification in wild blueberry fields
publisher KeAi Communications Co., Ltd.
series Artificial Intelligence in Agriculture
issn 2589-7217
publishDate 2021-01-01
description A Gabor wavelets based technique was investigated as a potential tool for scene classification (into one of bare patch, plant, or weed) for its ultimate utility in site-specific application of agrochemicals in wild blueberry fields.Images were gathered from five sites located in central Nova Scotia, Canada. Gabor wavelet features extracted from these images were used to classify scenes according to visually determined classes using step-wise linear discriminant analysis.For individual fields, classification accuracy attained ranged between 87.9% and 98.3%; selected Gabor features ranged between 27 and 72; contextual accuracy for herbicide ranged between 67.5% and 96.7%, and contextual accuracy for fertilizer ranged between 63.6% and 97.1%. The pooled scenes yielded a classification accuracy of 81.4%, and contextual accuracy figures of 61.1% and 73.1% for herbicide and fertilizer, respectively, with selected Gabor features of 36.Calibrations based on LDA coefficients from the pooled scenes could help avoid the need to re-calibrate for each field, whereas those based on individual field LDA coefficients could improve accuracy, hence enable saving on expensive agrochemicals.
topic Wavelet transforms
Wild blueberry
Computer vision
Machine learning
Discriminant analysis
Precision agriculture
url http://www.sciencedirect.com/science/article/pii/S2589721721000155
work_keys_str_mv AT gashawayalew aninvestigationintothepotentialofgaborwaveletfeaturesforsceneclassificationinwildblueberryfields
AT qamaruzzaman aninvestigationintothepotentialofgaborwaveletfeaturesforsceneclassificationinwildblueberryfields
AT arnoldwschumann aninvestigationintothepotentialofgaborwaveletfeaturesforsceneclassificationinwildblueberryfields
AT davidcpercival aninvestigationintothepotentialofgaborwaveletfeaturesforsceneclassificationinwildblueberryfields
AT youngkichang aninvestigationintothepotentialofgaborwaveletfeaturesforsceneclassificationinwildblueberryfields
AT gashawayalew investigationintothepotentialofgaborwaveletfeaturesforsceneclassificationinwildblueberryfields
AT qamaruzzaman investigationintothepotentialofgaborwaveletfeaturesforsceneclassificationinwildblueberryfields
AT arnoldwschumann investigationintothepotentialofgaborwaveletfeaturesforsceneclassificationinwildblueberryfields
AT davidcpercival investigationintothepotentialofgaborwaveletfeaturesforsceneclassificationinwildblueberryfields
AT youngkichang investigationintothepotentialofgaborwaveletfeaturesforsceneclassificationinwildblueberryfields
_version_ 1721523165986816000