沈阳农业大学学报2025，Vol.56Issue(6)：10-19,10.DOI:10.3969/j.issn.1000-1700.2025.06.002
基于PIOSL模型的水稻叶片叶绿素含量反演方法研究
Research on Inversion Method of Chlorophyll Content in Rice Leaves Based on PIOSL Model
摘要
Abstract
[Objective]Chlorophyll content is an important indicator for assessing the health of rice plants.Precise monitoring of chlorophyll content in rice leaves is of vital importance.Traditional methods for detecting chlorophyll content in rice leaves require destructive sampling,making it difficult to obtain real-time dynamic changes.With the rapid development of information technology,hyperspectral data combined with machine learning methods can be used to retrieve chlorophyll content in rice leaves.[Methods]Based on the hyperspectral simulation data of leaves output by the PIOSL(PROSPECT considering the internal optical structure of the leaves)model,three methods,namely Successive Projections Algorithm(SPA),Competitive Adaptive Reweighted Sampling(CARS),and Random Frog(RF),were adopted for feature band selection.Then,three algorithms,Particle Swarm Optimization(PSO),Whale Optimization Algorithm(WOA),and Love Evolution Algorithm(LEA),were used to optimize the Extreme Learning Machine(ELM)for the inversion of chlorophyll content in rice leaves.[Results]The research results show that based on the simulated data output from the PIOSL model,using three feature band selection methods,CARS,SPA,and RF,were used to select 19,9,and 10 feature bands respectively,ELMs optimized by PSO,WOA,and LEA were created.By comparing the performance of the nine models,it was concluded that the CARS-LEA-ELM model had the best effect in inverting the chlorophyll content of rice leaves,with a test set R2of 0.929 and RMSE of 4.116 μg·cm-2.[Conclusion]Through the comprehensive evaluation of the inversion results,the optimal inversion model construction scheme was determined to achieve high-precision inversion of the chlorophyll content of rice leaves and provide support for field management.关键词
辐射传输/高光谱/叶绿素含量/反演/水稻Key words
radiative transfer/hyperspectral/chlorophyll content/retrieval/rice分类
农业科技引用本文复制引用
于丰华,刘蕊,金忠煜,相爽,齐鑫,周久琳,李世隆..基于PIOSL模型的水稻叶片叶绿素含量反演方法研究[J].沈阳农业大学学报,2025,56(6):10-19,10.基金项目
辽宁省教育厅平台项目(JYTPT2024002) （JYTPT2024002）
