Estimation of corn yield based on hyperspectral imagery and convolutional neural network

•Characteristics of hyperspectral and color images were used to classify corn yield.•Training and testing were conducted with data collected from a corn field.•Accuracy of different types of CNN models were investigated.•Corn yield estimates using CNN provided insights for improving yield. Corn is a...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Computers and electronics in agriculture 2021-05, Vol.184, p.106092, Article 106092
Hauptverfasser: Yang, Wei, Nigon, Tyler, Hao, Ziyuan, Dias Paiao, Gabriel, Fernández, Fabián G., Mulla, David, Yang, Ce
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page 106092
container_title Computers and electronics in agriculture
container_volume 184
creator Yang, Wei
Nigon, Tyler
Hao, Ziyuan
Dias Paiao, Gabriel
Fernández, Fabián G.
Mulla, David
Yang, Ce
description •Characteristics of hyperspectral and color images were used to classify corn yield.•Training and testing were conducted with data collected from a corn field.•Accuracy of different types of CNN models were investigated.•Corn yield estimates using CNN provided insights for improving yield. Corn is an important food crop in the world, widely distributed in many countries because of its excellent environmental adaptability. Moreover, corn is an important feed source for animal production and it is an indispensable raw material for many different industries. With increasing human population and decreasing arable land, there is an increased focus on increasing yield of corn. Convolutional neural network (CNN) analysis can be used for non-destructive yield prediction and is well suited for classification and feature extraction. The overall objective of this experiment was to use hyperspectral imagery to train a CNN classification model to estimate corn grain yield. High resolution hyperspectral imagery was captured at five corn growth stages - V5 (five leaves with visible leaf collars), V8 (eight leaves with visible leaf collars), V10 (ten leaves with visible leaf collars), V12 (12 leaves with visible leaf collars), and R2 (blister stage). Hyperspectral imagery was denoised using the wavelet analysis method, then was used to train and validate the CNN model. The spectral information reflecting the internal characteristics and the spatial information provided by the color image (red, green and blue bands extracted from hyperspectral image) reflecting the external characteristics of corn growth are extracted for modelling and verification. The results show that the spectral and color image-based integrated CNN model has a classification accuracy of 75.50%. In contrast, the accuracy of a one-dimensional CNN model based only on spectral information or a two-dimensional CNN model based only on color image information were 60.39% and 32.17%, respectively. The integrated CNN model (spectral information plus color image information) is better than results of the individual one-dimensional CNN or two-dimensional CNN models. In addition. The Kappa coefficient of integrated CNN model is 0.69, which indicates a high consistency of classification. Comprehensive use of spectral information and color image information, which represent information about the inner and outer corn canopy can provide more accurate corn yield prediction than one-dimensional or two-dimensional CNN m
doi_str_mv 10.1016/j.compag.2021.106092
format Article
fullrecord <record><control><sourceid>proquest_cross</sourceid><recordid>TN_cdi_proquest_journals_2528883537</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><els_id>S0168169921001101</els_id><sourcerecordid>2528883537</sourcerecordid><originalsourceid>FETCH-LOGICAL-c380t-3a607da0fc677bd71c17d941e09c4449a344c686445fd304a99286c22670c1ae3</originalsourceid><addsrcrecordid>eNp9kE1LxDAQhoMouK7-Aw8Fz12TNJuPiyDL-gELXhS8hWwyXVu7TU3alf57s9azpyGT5x1mHoSuCV4QTPhtvbB-35ndgmJKUotjRU_QjEhBc0GwOEWzhMmccKXO0UWMNU5vJcUMva9jX-1NX_k282VmfWizsYLGZVsTwWWp_TF2EGIHtg-myRK8gzBmpnWJbg--GY7h9NPCEH5L_-3D5yU6K00T4eqvztHbw_p19ZRvXh6fV_eb3BYS93lhOBbO4NJyIbZOEEuEU4wAVpYxpkzBmOWSM7YsXYGZUYpKbinlAltioJijm2luF_zXALHXtR9C2idquqRSymJZiESxibLBxxig1F1Il4RRE6yPDnWtJ4f66FBPDlPsbopBuuBQQdDRVtBacFVIPrTz1f8DfgD6H3xi</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2528883537</pqid></control><display><type>article</type><title>Estimation of corn yield based on hyperspectral imagery and convolutional neural network</title><source>Access via ScienceDirect (Elsevier)</source><creator>Yang, Wei ; Nigon, Tyler ; Hao, Ziyuan ; Dias Paiao, Gabriel ; Fernández, Fabián G. ; Mulla, David ; Yang, Ce</creator><creatorcontrib>Yang, Wei ; Nigon, Tyler ; Hao, Ziyuan ; Dias Paiao, Gabriel ; Fernández, Fabián G. ; Mulla, David ; Yang, Ce</creatorcontrib><description>•Characteristics of hyperspectral and color images were used to classify corn yield.•Training and testing were conducted with data collected from a corn field.•Accuracy of different types of CNN models were investigated.•Corn yield estimates using CNN provided insights for improving yield. Corn is an important food crop in the world, widely distributed in many countries because of its excellent environmental adaptability. Moreover, corn is an important feed source for animal production and it is an indispensable raw material for many different industries. With increasing human population and decreasing arable land, there is an increased focus on increasing yield of corn. Convolutional neural network (CNN) analysis can be used for non-destructive yield prediction and is well suited for classification and feature extraction. The overall objective of this experiment was to use hyperspectral imagery to train a CNN classification model to estimate corn grain yield. High resolution hyperspectral imagery was captured at five corn growth stages - V5 (five leaves with visible leaf collars), V8 (eight leaves with visible leaf collars), V10 (ten leaves with visible leaf collars), V12 (12 leaves with visible leaf collars), and R2 (blister stage). Hyperspectral imagery was denoised using the wavelet analysis method, then was used to train and validate the CNN model. The spectral information reflecting the internal characteristics and the spatial information provided by the color image (red, green and blue bands extracted from hyperspectral image) reflecting the external characteristics of corn growth are extracted for modelling and verification. The results show that the spectral and color image-based integrated CNN model has a classification accuracy of 75.50%. In contrast, the accuracy of a one-dimensional CNN model based only on spectral information or a two-dimensional CNN model based only on color image information were 60.39% and 32.17%, respectively. The integrated CNN model (spectral information plus color image information) is better than results of the individual one-dimensional CNN or two-dimensional CNN models. In addition. The Kappa coefficient of integrated CNN model is 0.69, which indicates a high consistency of classification. Comprehensive use of spectral information and color image information, which represent information about the inner and outer corn canopy can provide more accurate corn yield prediction than one-dimensional or two-dimensional CNN models.</description><identifier>ISSN: 0168-1699</identifier><identifier>EISSN: 1872-7107</identifier><identifier>DOI: 10.1016/j.compag.2021.106092</identifier><language>eng</language><publisher>Amsterdam: Elsevier B.V</publisher><subject>Arable land ; Artificial neural networks ; Blistering ; Classification ; CNN ; Collars ; Color imagery ; Corn yield ; Crop yield ; Deep learning theory ; Feature extraction ; Hyperspectral Image ; Hyperspectral imaging ; Image classification ; Image resolution ; Leaves ; Model accuracy ; Neural networks ; Noise reduction ; Spatial data ; Spectra ; Two dimensional models ; Wavelet analysis</subject><ispartof>Computers and electronics in agriculture, 2021-05, Vol.184, p.106092, Article 106092</ispartof><rights>2021 Elsevier B.V.</rights><rights>Copyright Elsevier BV May 2021</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c380t-3a607da0fc677bd71c17d941e09c4449a344c686445fd304a99286c22670c1ae3</citedby><cites>FETCH-LOGICAL-c380t-3a607da0fc677bd71c17d941e09c4449a344c686445fd304a99286c22670c1ae3</cites></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://dx.doi.org/10.1016/j.compag.2021.106092$$EHTML$$P50$$Gelsevier$$H</linktohtml><link.rule.ids>314,780,784,3550,27924,27925,45995</link.rule.ids></links><search><creatorcontrib>Yang, Wei</creatorcontrib><creatorcontrib>Nigon, Tyler</creatorcontrib><creatorcontrib>Hao, Ziyuan</creatorcontrib><creatorcontrib>Dias Paiao, Gabriel</creatorcontrib><creatorcontrib>Fernández, Fabián G.</creatorcontrib><creatorcontrib>Mulla, David</creatorcontrib><creatorcontrib>Yang, Ce</creatorcontrib><title>Estimation of corn yield based on hyperspectral imagery and convolutional neural network</title><title>Computers and electronics in agriculture</title><description>•Characteristics of hyperspectral and color images were used to classify corn yield.•Training and testing were conducted with data collected from a corn field.•Accuracy of different types of CNN models were investigated.•Corn yield estimates using CNN provided insights for improving yield. Corn is an important food crop in the world, widely distributed in many countries because of its excellent environmental adaptability. Moreover, corn is an important feed source for animal production and it is an indispensable raw material for many different industries. With increasing human population and decreasing arable land, there is an increased focus on increasing yield of corn. Convolutional neural network (CNN) analysis can be used for non-destructive yield prediction and is well suited for classification and feature extraction. The overall objective of this experiment was to use hyperspectral imagery to train a CNN classification model to estimate corn grain yield. High resolution hyperspectral imagery was captured at five corn growth stages - V5 (five leaves with visible leaf collars), V8 (eight leaves with visible leaf collars), V10 (ten leaves with visible leaf collars), V12 (12 leaves with visible leaf collars), and R2 (blister stage). Hyperspectral imagery was denoised using the wavelet analysis method, then was used to train and validate the CNN model. The spectral information reflecting the internal characteristics and the spatial information provided by the color image (red, green and blue bands extracted from hyperspectral image) reflecting the external characteristics of corn growth are extracted for modelling and verification. The results show that the spectral and color image-based integrated CNN model has a classification accuracy of 75.50%. In contrast, the accuracy of a one-dimensional CNN model based only on spectral information or a two-dimensional CNN model based only on color image information were 60.39% and 32.17%, respectively. The integrated CNN model (spectral information plus color image information) is better than results of the individual one-dimensional CNN or two-dimensional CNN models. In addition. The Kappa coefficient of integrated CNN model is 0.69, which indicates a high consistency of classification. Comprehensive use of spectral information and color image information, which represent information about the inner and outer corn canopy can provide more accurate corn yield prediction than one-dimensional or two-dimensional CNN models.</description><subject>Arable land</subject><subject>Artificial neural networks</subject><subject>Blistering</subject><subject>Classification</subject><subject>CNN</subject><subject>Collars</subject><subject>Color imagery</subject><subject>Corn yield</subject><subject>Crop yield</subject><subject>Deep learning theory</subject><subject>Feature extraction</subject><subject>Hyperspectral Image</subject><subject>Hyperspectral imaging</subject><subject>Image classification</subject><subject>Image resolution</subject><subject>Leaves</subject><subject>Model accuracy</subject><subject>Neural networks</subject><subject>Noise reduction</subject><subject>Spatial data</subject><subject>Spectra</subject><subject>Two dimensional models</subject><subject>Wavelet analysis</subject><issn>0168-1699</issn><issn>1872-7107</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2021</creationdate><recordtype>article</recordtype><recordid>eNp9kE1LxDAQhoMouK7-Aw8Fz12TNJuPiyDL-gELXhS8hWwyXVu7TU3alf57s9azpyGT5x1mHoSuCV4QTPhtvbB-35ndgmJKUotjRU_QjEhBc0GwOEWzhMmccKXO0UWMNU5vJcUMva9jX-1NX_k282VmfWizsYLGZVsTwWWp_TF2EGIHtg-myRK8gzBmpnWJbg--GY7h9NPCEH5L_-3D5yU6K00T4eqvztHbw_p19ZRvXh6fV_eb3BYS93lhOBbO4NJyIbZOEEuEU4wAVpYxpkzBmOWSM7YsXYGZUYpKbinlAltioJijm2luF_zXALHXtR9C2idquqRSymJZiESxibLBxxig1F1Il4RRE6yPDnWtJ4f66FBPDlPsbopBuuBQQdDRVtBacFVIPrTz1f8DfgD6H3xi</recordid><startdate>202105</startdate><enddate>202105</enddate><creator>Yang, Wei</creator><creator>Nigon, Tyler</creator><creator>Hao, Ziyuan</creator><creator>Dias Paiao, Gabriel</creator><creator>Fernández, Fabián G.</creator><creator>Mulla, David</creator><creator>Yang, Ce</creator><general>Elsevier B.V</general><general>Elsevier BV</general><scope>AAYXX</scope><scope>CITATION</scope><scope>7SC</scope><scope>7SP</scope><scope>8FD</scope><scope>FR3</scope><scope>JQ2</scope><scope>KR7</scope><scope>L7M</scope><scope>L~C</scope><scope>L~D</scope></search><sort><creationdate>202105</creationdate><title>Estimation of corn yield based on hyperspectral imagery and convolutional neural network</title><author>Yang, Wei ; Nigon, Tyler ; Hao, Ziyuan ; Dias Paiao, Gabriel ; Fernández, Fabián G. ; Mulla, David ; Yang, Ce</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c380t-3a607da0fc677bd71c17d941e09c4449a344c686445fd304a99286c22670c1ae3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2021</creationdate><topic>Arable land</topic><topic>Artificial neural networks</topic><topic>Blistering</topic><topic>Classification</topic><topic>CNN</topic><topic>Collars</topic><topic>Color imagery</topic><topic>Corn yield</topic><topic>Crop yield</topic><topic>Deep learning theory</topic><topic>Feature extraction</topic><topic>Hyperspectral Image</topic><topic>Hyperspectral imaging</topic><topic>Image classification</topic><topic>Image resolution</topic><topic>Leaves</topic><topic>Model accuracy</topic><topic>Neural networks</topic><topic>Noise reduction</topic><topic>Spatial data</topic><topic>Spectra</topic><topic>Two dimensional models</topic><topic>Wavelet analysis</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Yang, Wei</creatorcontrib><creatorcontrib>Nigon, Tyler</creatorcontrib><creatorcontrib>Hao, Ziyuan</creatorcontrib><creatorcontrib>Dias Paiao, Gabriel</creatorcontrib><creatorcontrib>Fernández, Fabián G.</creatorcontrib><creatorcontrib>Mulla, David</creatorcontrib><creatorcontrib>Yang, Ce</creatorcontrib><collection>CrossRef</collection><collection>Computer and Information Systems Abstracts</collection><collection>Electronics &amp; Communications Abstracts</collection><collection>Technology Research Database</collection><collection>Engineering Research Database</collection><collection>ProQuest Computer Science Collection</collection><collection>Civil Engineering Abstracts</collection><collection>Advanced Technologies Database with Aerospace</collection><collection>Computer and Information Systems Abstracts – Academic</collection><collection>Computer and Information Systems Abstracts Professional</collection><jtitle>Computers and electronics in agriculture</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Yang, Wei</au><au>Nigon, Tyler</au><au>Hao, Ziyuan</au><au>Dias Paiao, Gabriel</au><au>Fernández, Fabián G.</au><au>Mulla, David</au><au>Yang, Ce</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Estimation of corn yield based on hyperspectral imagery and convolutional neural network</atitle><jtitle>Computers and electronics in agriculture</jtitle><date>2021-05</date><risdate>2021</risdate><volume>184</volume><spage>106092</spage><pages>106092-</pages><artnum>106092</artnum><issn>0168-1699</issn><eissn>1872-7107</eissn><abstract>•Characteristics of hyperspectral and color images were used to classify corn yield.•Training and testing were conducted with data collected from a corn field.•Accuracy of different types of CNN models were investigated.•Corn yield estimates using CNN provided insights for improving yield. Corn is an important food crop in the world, widely distributed in many countries because of its excellent environmental adaptability. Moreover, corn is an important feed source for animal production and it is an indispensable raw material for many different industries. With increasing human population and decreasing arable land, there is an increased focus on increasing yield of corn. Convolutional neural network (CNN) analysis can be used for non-destructive yield prediction and is well suited for classification and feature extraction. The overall objective of this experiment was to use hyperspectral imagery to train a CNN classification model to estimate corn grain yield. High resolution hyperspectral imagery was captured at five corn growth stages - V5 (five leaves with visible leaf collars), V8 (eight leaves with visible leaf collars), V10 (ten leaves with visible leaf collars), V12 (12 leaves with visible leaf collars), and R2 (blister stage). Hyperspectral imagery was denoised using the wavelet analysis method, then was used to train and validate the CNN model. The spectral information reflecting the internal characteristics and the spatial information provided by the color image (red, green and blue bands extracted from hyperspectral image) reflecting the external characteristics of corn growth are extracted for modelling and verification. The results show that the spectral and color image-based integrated CNN model has a classification accuracy of 75.50%. In contrast, the accuracy of a one-dimensional CNN model based only on spectral information or a two-dimensional CNN model based only on color image information were 60.39% and 32.17%, respectively. The integrated CNN model (spectral information plus color image information) is better than results of the individual one-dimensional CNN or two-dimensional CNN models. In addition. The Kappa coefficient of integrated CNN model is 0.69, which indicates a high consistency of classification. Comprehensive use of spectral information and color image information, which represent information about the inner and outer corn canopy can provide more accurate corn yield prediction than one-dimensional or two-dimensional CNN models.</abstract><cop>Amsterdam</cop><pub>Elsevier B.V</pub><doi>10.1016/j.compag.2021.106092</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier ISSN: 0168-1699
ispartof Computers and electronics in agriculture, 2021-05, Vol.184, p.106092, Article 106092
issn 0168-1699
1872-7107
language eng
recordid cdi_proquest_journals_2528883537
source Access via ScienceDirect (Elsevier)
subjects Arable land
Artificial neural networks
Blistering
Classification
CNN
Collars
Color imagery
Corn yield
Crop yield
Deep learning theory
Feature extraction
Hyperspectral Image
Hyperspectral imaging
Image classification
Image resolution
Leaves
Model accuracy
Neural networks
Noise reduction
Spatial data
Spectra
Two dimensional models
Wavelet analysis
title Estimation of corn yield based on hyperspectral imagery and convolutional neural network
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-23T03%3A53%3A16IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_cross&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Estimation%20of%20corn%20yield%20based%20on%20hyperspectral%20imagery%20and%20convolutional%20neural%20network&rft.jtitle=Computers%20and%20electronics%20in%20agriculture&rft.au=Yang,%20Wei&rft.date=2021-05&rft.volume=184&rft.spage=106092&rft.pages=106092-&rft.artnum=106092&rft.issn=0168-1699&rft.eissn=1872-7107&rft_id=info:doi/10.1016/j.compag.2021.106092&rft_dat=%3Cproquest_cross%3E2528883537%3C/proquest_cross%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2528883537&rft_id=info:pmid/&rft_els_id=S0168169921001101&rfr_iscdi=true