{"title":"Crop chlorophyll detection based on multiexcitation fluorescence imaging analysis","authors":"","doi":"10.1016/j.biosystemseng.2024.07.012","DOIUrl":null,"url":null,"abstract":"<div><p>The chlorophyll content of wheat was assessed using multispectral fluorescence imaging (MSFI). Ultraviolet (UV) light (365 nm)-induced fluorescence images at 440, 520, 690, and 740 nm, and visible light (460, and 610 nm)-induced fluorescence images at 690 and 740 nm were acquired while leaf chlorophyll content was measured using SPAD 520. The fluorescence images were processed after segmentation and channel extraction to calculate the parameters of each leaf based on fluorescence images (<span><math><mrow><msub><mi>F</mi><mi>u</mi></msub><mn>440</mn></mrow></math></span>, <span><math><mrow><msub><mi>F</mi><mi>u</mi></msub><mn>520</mn></mrow></math></span>, <span><math><mrow><msub><mi>F</mi><mi>u</mi></msub><mn>690</mn></mrow></math></span>, and <span><math><mrow><msub><mi>F</mi><mi>u</mi></msub><mn>740</mn></mrow></math></span>) obtained by UV excitation, and fluorescence images (<span><math><mrow><msub><mi>F</mi><mi>u</mi></msub><mn>440</mn></mrow></math></span>, <span><math><mrow><msub><mi>F</mi><mi>u</mi></msub><mn>520</mn></mrow></math></span>, <span><math><mrow><msub><mi>F</mi><mi>u</mi></msub><mn>690</mn></mrow></math></span>, <span><math><mrow><msub><mi>F</mi><mi>u</mi></msub><mn>740</mn></mrow></math></span>, <span><math><mrow><msub><mi>F</mi><mi>b</mi></msub><mn>690</mn></mrow></math></span>, <span><math><mrow><msub><mi>F</mi><mi>b</mi></msub><mn>740</mn></mrow></math></span>, and <span><math><mrow><msub><mi>F</mi><mi>r</mi></msub><mn>740</mn></mrow></math></span>) obtained by three excitations of 365 nm, 460 nm, and 610 nm light. 12 fluorescence ratio parameters under UV excitation and 26 fluorescence ratio parameters under three excitations were calculated. The correlation analysis revealed that the fluorescence parameters (<span><math><mrow><msub><mi>F</mi><mi>r</mi></msub><mn>740</mn></mrow></math></span>, <span><math><mrow><msub><mi>F</mi><mi>u</mi></msub><mn>440</mn></mrow></math></span>, <span><math><mrow><msub><mi>F</mi><mi>u</mi></msub><mn>520</mn></mrow></math></span>, <span><math><mrow><msub><mi>F</mi><mi>u</mi></msub><mn>690</mn></mrow></math></span>, <span><math><mrow><msub><mi>F</mi><mi>u</mi></msub><mn>740</mn></mrow></math></span>, <span><math><mrow><msub><mi>F</mi><mi>b</mi></msub><mn>690</mn></mrow></math></span>, <span><math><mrow><msub><mi>F</mi><mi>b</mi></msub><mn>740</mn></mrow></math></span>, <span><math><mrow><msub><mi>F</mi><mi>u</mi></msub><mn>440</mn><mo>/</mo><msub><mi>F</mi><mi>u</mi></msub><mn>520</mn></mrow></math></span>, <span><math><mrow><msub><mi>F</mi><mi>u</mi></msub><mn>520</mn><mo>/</mo><msub><mi>F</mi><mi>u</mi></msub><mn>690</mn></mrow></math></span>, and <span><math><mrow><msub><mi>F</mi><mi>u</mi></msub><mn>740</mn><mo>/</mo><msub><mi>F</mi><mi>r</mi></msub><mn>740</mn></mrow></math></span>) showed a strong correlation with the chlorophyll content. These parameters have the potential to measure the chlorophyll content. Subsequently, stepwise regression analysis (SRA) was employed to screen 16 fluorescence parameters under UV excitation and 33 fluorescence parameters under three excitations, with the objective of identifying and eliminating redundant variables. Finally, four variables (<span><math><mrow><msub><mi>F</mi><mi>u</mi></msub><mn>520</mn></mrow></math></span>, <span><math><mrow><msub><mi>F</mi><mi>u</mi></msub><mn>690</mn></mrow></math></span>, <span><math><mrow><msub><mi>F</mi><mi>u</mi></msub><mn>740</mn></mrow></math></span>, and <span><math><mrow><msub><mi>F</mi><mi>u</mi></msub><mn>690</mn><mo>/</mo><msub><mi>F</mi><mi>u</mi></msub><mn>520</mn></mrow></math></span>) under UV excitation and five variables (<span><math><mrow><msub><mi>F</mi><mi>r</mi></msub><mn>740</mn></mrow></math></span>, <span><math><mrow><msub><mi>F</mi><mi>u</mi></msub><mn>520</mn></mrow></math></span>, <span><math><mrow><msub><mi>F</mi><mi>b</mi></msub><mn>740</mn></mrow></math></span>, <span><math><mrow><msub><mi>F</mi><mi>u</mi></msub><mn>740</mn><mo>/</mo><msub><mi>F</mi><mi>u</mi></msub><mn>690</mn></mrow></math></span>, and <span><math><mrow><msub><mi>F</mi><mi>b</mi></msub><mn>740</mn><mo>/</mo><msub><mi>F</mi><mi>b</mi></msub><mn>690</mn></mrow></math></span>) under three excitations were selected. The partial least squares regression (PLSR) model, constructed using three excitations, demonstrated enhanced performance with an <span><math><mrow><msubsup><mi>R</mi><mi>c</mi><mn>2</mn></msubsup></mrow></math></span> of 0.901, <span><math><mrow><msubsup><mi>R</mi><mi>v</mi><mn>2</mn></msubsup></mrow></math></span> of 0.904, root mean square error (RMSE) of calibration of 4.398, and RMSE of validation of 4.267. Multiexcitation fluorescence based on three excitations techniques has better performance for evaluating chlorophyll content.</p></div>","PeriodicalId":9173,"journal":{"name":"Biosystems Engineering","volume":null,"pages":null},"PeriodicalIF":4.4000,"publicationDate":"2024-07-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Biosystems Engineering","FirstCategoryId":"97","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S153751102400165X","RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"AGRICULTURAL ENGINEERING","Score":null,"Total":0}
引用次数: 0
Abstract
The chlorophyll content of wheat was assessed using multispectral fluorescence imaging (MSFI). Ultraviolet (UV) light (365 nm)-induced fluorescence images at 440, 520, 690, and 740 nm, and visible light (460, and 610 nm)-induced fluorescence images at 690 and 740 nm were acquired while leaf chlorophyll content was measured using SPAD 520. The fluorescence images were processed after segmentation and channel extraction to calculate the parameters of each leaf based on fluorescence images (, , , and ) obtained by UV excitation, and fluorescence images (, , , , , , and ) obtained by three excitations of 365 nm, 460 nm, and 610 nm light. 12 fluorescence ratio parameters under UV excitation and 26 fluorescence ratio parameters under three excitations were calculated. The correlation analysis revealed that the fluorescence parameters (, , , , , , , , , and ) showed a strong correlation with the chlorophyll content. These parameters have the potential to measure the chlorophyll content. Subsequently, stepwise regression analysis (SRA) was employed to screen 16 fluorescence parameters under UV excitation and 33 fluorescence parameters under three excitations, with the objective of identifying and eliminating redundant variables. Finally, four variables (, , , and ) under UV excitation and five variables (, , , , and ) under three excitations were selected. The partial least squares regression (PLSR) model, constructed using three excitations, demonstrated enhanced performance with an of 0.901, of 0.904, root mean square error (RMSE) of calibration of 4.398, and RMSE of validation of 4.267. Multiexcitation fluorescence based on three excitations techniques has better performance for evaluating chlorophyll content.
期刊介绍:
Biosystems Engineering publishes research in engineering and the physical sciences that represent advances in understanding or modelling of the performance of biological systems for sustainable developments in land use and the environment, agriculture and amenity, bioproduction processes and the food chain. The subject matter of the journal reflects the wide range and interdisciplinary nature of research in engineering for biological systems.