Alfalfa detection and stem count from proximal images using a combination of deep neural networks and machine learning

IF 7.7 1区 农林科学 Q1 AGRICULTURE, MULTIDISCIPLINARY Computers and Electronics in Agriculture Pub Date : 2025-02-16 DOI:10.1016/j.compag.2025.110115
Hazhir Bahrami , Karem Chokmani , Saeid Homayouni , Viacheslav I. Adamchuk , Md Saifuzzaman , Maxime Leduc
{"title":"Alfalfa detection and stem count from proximal images using a combination of deep neural networks and machine learning","authors":"Hazhir Bahrami ,&nbsp;Karem Chokmani ,&nbsp;Saeid Homayouni ,&nbsp;Viacheslav I. Adamchuk ,&nbsp;Md Saifuzzaman ,&nbsp;Maxime Leduc","doi":"10.1016/j.compag.2025.110115","DOIUrl":null,"url":null,"abstract":"<div><div>Among various types of forages, Alfalfa (Medicago sativa) is a crucial forage crop that plays a vital role in livestock nutrition and sustainable agriculture. As a result of its ability to adapt to different weather conditions and its high nitrogen fixation capability, this crop produces high-quality forage that contains between 15 and 22 % protein. It is fortunately possible to improve the overall prediction of forage biomass and quality prior to harvest through remote sensing technologies. The recent advent of deep Convolution Neural Networks (deep CNNs) enables researchers to utilize these incredible algorithms. This study aims to build a model to count the number of alfalfa stems from proximal images. To this end, we first utilized a deep CNN encoder-decoder to segment alfalfa and other background objects in a field, such as soil and grass. Subsequently, we employed the alfalfa cover fractions derived from the proximal images to develop and train machine learning regression models for estimating the stem count in the images. This study uses many proximal images taken from significant number of fields in four provinces of Canada over three consecutive years. A combination of real and synthetic images has been utilized to feed the deep neural network encoder-decoder. This study gathered roughly 3447 alfalfa images, 5332 grass images, and 9241 background images for training the encoder-decoder model. With data augmentation, we prepared about 60,000 annotated images of alfalfa fields containing alfalfa, grass, and background utilizing a pre-trained model in less than an hour. Several convolutional neural network encoder-decoder models have also been utilized in this study. Simple U-Net, Attention U-Net (Att U-Net), and ResU-Net with attention gates have been trained to detect alfalfa and differentiate it from other objects. The best Intersections over Union (IoU) for simple U-Net classes were 0.98, 0.93, and 0.80 for background, alfalfa and grass, respectively. Simple U-Net with synthetic data provides a promising result over unseen real images and requires an RGB iPad image for field-specific alfalfa detection. It was also observed that simple U-Net has slightly better accuracy than attention U-Net and attention ResU-Net. Finally, we built regression models between the alfalfa cover fraction in the original images taken by iPad, and the mean alfalfa stems per square foot. Random forest (RF), Support Vector Regression (SVR), and Extreme Gradient Boosting (XGB) methods have been utilized to estimate the number of stems in the images. RF was the best model for estimating the number of alfalfa stems relative to other machine learning algorithms, with a coefficient of determination (R<sup>2</sup>) of 0.82, root-mean-square error of 13.00, and mean absolute error of 10.07.</div></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":"232 ","pages":"Article 110115"},"PeriodicalIF":7.7000,"publicationDate":"2025-02-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Computers and Electronics in Agriculture","FirstCategoryId":"97","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0168169925002212","RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"AGRICULTURE, MULTIDISCIPLINARY","Score":null,"Total":0}
引用次数: 0

Abstract

Among various types of forages, Alfalfa (Medicago sativa) is a crucial forage crop that plays a vital role in livestock nutrition and sustainable agriculture. As a result of its ability to adapt to different weather conditions and its high nitrogen fixation capability, this crop produces high-quality forage that contains between 15 and 22 % protein. It is fortunately possible to improve the overall prediction of forage biomass and quality prior to harvest through remote sensing technologies. The recent advent of deep Convolution Neural Networks (deep CNNs) enables researchers to utilize these incredible algorithms. This study aims to build a model to count the number of alfalfa stems from proximal images. To this end, we first utilized a deep CNN encoder-decoder to segment alfalfa and other background objects in a field, such as soil and grass. Subsequently, we employed the alfalfa cover fractions derived from the proximal images to develop and train machine learning regression models for estimating the stem count in the images. This study uses many proximal images taken from significant number of fields in four provinces of Canada over three consecutive years. A combination of real and synthetic images has been utilized to feed the deep neural network encoder-decoder. This study gathered roughly 3447 alfalfa images, 5332 grass images, and 9241 background images for training the encoder-decoder model. With data augmentation, we prepared about 60,000 annotated images of alfalfa fields containing alfalfa, grass, and background utilizing a pre-trained model in less than an hour. Several convolutional neural network encoder-decoder models have also been utilized in this study. Simple U-Net, Attention U-Net (Att U-Net), and ResU-Net with attention gates have been trained to detect alfalfa and differentiate it from other objects. The best Intersections over Union (IoU) for simple U-Net classes were 0.98, 0.93, and 0.80 for background, alfalfa and grass, respectively. Simple U-Net with synthetic data provides a promising result over unseen real images and requires an RGB iPad image for field-specific alfalfa detection. It was also observed that simple U-Net has slightly better accuracy than attention U-Net and attention ResU-Net. Finally, we built regression models between the alfalfa cover fraction in the original images taken by iPad, and the mean alfalfa stems per square foot. Random forest (RF), Support Vector Regression (SVR), and Extreme Gradient Boosting (XGB) methods have been utilized to estimate the number of stems in the images. RF was the best model for estimating the number of alfalfa stems relative to other machine learning algorithms, with a coefficient of determination (R2) of 0.82, root-mean-square error of 13.00, and mean absolute error of 10.07.
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
求助全文
约1分钟内获得全文 去求助
来源期刊
Computers and Electronics in Agriculture
Computers and Electronics in Agriculture 工程技术-计算机:跨学科应用
CiteScore
15.30
自引率
14.50%
发文量
800
审稿时长
62 days
期刊介绍: Computers and Electronics in Agriculture provides international coverage of advancements in computer hardware, software, electronic instrumentation, and control systems applied to agricultural challenges. Encompassing agronomy, horticulture, forestry, aquaculture, and animal farming, the journal publishes original papers, reviews, and applications notes. It explores the use of computers and electronics in plant or animal agricultural production, covering topics like agricultural soils, water, pests, controlled environments, and waste. The scope extends to on-farm post-harvest operations and relevant technologies, including artificial intelligence, sensors, machine vision, robotics, networking, and simulation modeling. Its companion journal, Smart Agricultural Technology, continues the focus on smart applications in production agriculture.
期刊最新文献
Research on multi-layer model attitude recognition and picking strategy of small tomato picking robot Definition of a reference standard for performance evaluation of autonomous vehicles real-time obstacle detection and distance estimation in complex environments Printed RFID sensing system: The cost-effective way to IoT smart agriculture Real-time monitoring of fertilizer runoff at the watershed scale using a low-cost solar-powered Lego-like electrochemical water quality monitoring system AI-driven adaptive grasping and precise detaching robot for efficient citrus harvesting
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1