{"title":"Ground-based mobile imaging for detecting salt stress of cotton seedlings in the field","authors":"","doi":"10.1016/j.compag.2024.109550","DOIUrl":null,"url":null,"abstract":"<div><div>Salt stress is a one of the major abiotic stresses to cotton seedlings in Xinjiang. In practical situation, detecting salt stress in cotton seedling in the field often needs to consider the influence of other abiotic stresses, such as nitrogen deficiency, drought, drought and nitrogen deficiency combination. To achieve this goal, a ground-based optical sensing platform combining multicolor fluorescence with multispectral imaging with canopy height correction was developed to collect cotton seedlings images. The results showed that drought and nitrogen deficiency stress had similar effect with salt stress in plant coverage, plant height, and multispectral reflection characteristics. However, the combination of multicolor fluorescence and multispectral images could provide a powerful method for distinguishing them from each other. In the experiment, the multi-frequency image fusion network (MFIF-Net) based on the Laplacian pyramid outperformed wavelet transform and principal component-weighted averaging in image fusion. Ultimately, MFIF-Net-EfficientNet-b4 performed the best performance with overall accuracies of 89.03 % and 81.20 %, respectively for four (healthy, other stresses, sight salt stress and severe salt stress) and six categories (healthy, low nitrogen conditions, drought, drought and low nitrogen combination, slight salt stress, severe salt stress) with smaller resource requirements (parameter amount:14.67 M; FLOPs:3.91G). The results demonstrated the feasibility of MFIF-Net-EfficientNet-b4 coupled with ground-based optical sensing for detecting salt stress of cotton seedlings in the field.</div></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":null,"pages":null},"PeriodicalIF":7.7000,"publicationDate":"2024-10-17","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/S0168169924009414","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
Salt stress is a one of the major abiotic stresses to cotton seedlings in Xinjiang. In practical situation, detecting salt stress in cotton seedling in the field often needs to consider the influence of other abiotic stresses, such as nitrogen deficiency, drought, drought and nitrogen deficiency combination. To achieve this goal, a ground-based optical sensing platform combining multicolor fluorescence with multispectral imaging with canopy height correction was developed to collect cotton seedlings images. The results showed that drought and nitrogen deficiency stress had similar effect with salt stress in plant coverage, plant height, and multispectral reflection characteristics. However, the combination of multicolor fluorescence and multispectral images could provide a powerful method for distinguishing them from each other. In the experiment, the multi-frequency image fusion network (MFIF-Net) based on the Laplacian pyramid outperformed wavelet transform and principal component-weighted averaging in image fusion. Ultimately, MFIF-Net-EfficientNet-b4 performed the best performance with overall accuracies of 89.03 % and 81.20 %, respectively for four (healthy, other stresses, sight salt stress and severe salt stress) and six categories (healthy, low nitrogen conditions, drought, drought and low nitrogen combination, slight salt stress, severe salt stress) with smaller resource requirements (parameter amount:14.67 M; FLOPs:3.91G). The results demonstrated the feasibility of MFIF-Net-EfficientNet-b4 coupled with ground-based optical sensing for detecting salt stress of cotton seedlings in the field.
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.