中国农业科学 ›› 2025, Vol. 58 ›› Issue (7): 1321-1332.doi: 10.3864/j.issn.0578-1752.2025.07.006

• 耕作栽培·生理生化·农业信息技术 • 上一篇    下一篇

基于光谱指数的琯溪蜜柚叶片钙素含量估测模型研究

栗方亮(), 孔庆波(), 张青   

  1. 福建省农业科学院资源环境与土壤肥料研究所/福建省植物营养与肥料重点实验室,福州 350013
  • 收稿日期:2024-08-10 接受日期:2024-12-23 出版日期:2025-04-08 发布日期:2025-04-08
  • 通信作者:
    孔庆波,E-mail:
  • 联系方式: 栗方亮,E-mail:lifl007@qq.com。
  • 基金资助:
    福建省自然科学基金(2023J01362); 中央引导地方科技发展专项(2023L3022); 福建省农业科学院对外合作项目(DWHZ-2024-16)

Research on the Estimation Model of Calcium Content in Guanxi Honey Pomelo Leaves Based on Spectral Index

LI FangLiang(), KONG QingBo(), ZHANG Qing   

  1. Institute of Resources, Environment and Soil Fertilizer, Fujian Academy of Agricultural Sciences/Fujian Key Laboratory of Plant Nutrition and Fertilizer, Fuzhou 350013
  • Received:2024-08-10 Accepted:2024-12-23 Published:2025-04-08 Online:2025-04-08

摘要:

【目的】构建基于光谱分析的蜜柚叶片钙(Ca)素含量估测模型,可为蜜柚叶片钙素含量的监测和快速无损诊断提供理论基础。【方法】分析提取蜜柚叶片原始光谱及一阶导数光谱特征波段和光谱特征指数(差值光谱指数(DSI)、比值光谱指数(RSI)和归一化光谱指数(NDSI)),构建蜜柚叶片钙素含量的单变量估测模型、偏最小二乘估测模型(PLS)、反向传播神经网络估测模型(BPNN)、随机森林估测模型(RF)和支持向量机估测模型(SVM),并评价和验证蜜柚叶片钙素含量的最优光谱估测模型。【结果】蜜柚叶片原始光谱和一阶导数光谱与钙素含量存在多波段显著相关,基于原始光谱和一阶导数光谱的相关系数,最大的波长分别为553、714 nm和528、699、602 nm。基于原始光谱和一阶导数光谱与钙素含量相关性较显著的光谱指数分别为DSI790,1040、RSI910,990、NDSI900,990和NDSI′350,580、DSI′560,570、RSI′350,580。以RSI910,990、NDSI900,990、NDSI′350,580、DSI790,1040、DSI′560,570、RSI′350,580、DSI′528,602等光谱指数为自变量构建的多项式估测模型,决定系数R²较大(R²>0.60)。运用上述4种机器学习方法建立蜜柚叶片钙素含量高光谱估测模型,PLS、BPNN、RF和SVM 4种估测模型的R2分别为0.79、0.82、0.85和0.84,均方根误差(RMSE)分别为4.33、4.11、3.81和3.93,验证模型的R2分别为0.77、0.80、0.87和0.83,RMSE分别为4.50、4.28、3.67和3.90,估测模型的精确程度为RF>SVM>BPNN>PLS。【结论】蜜柚叶片钙素含量的4种模型进行精度对比分析表明,RF估测模型的预测性优于其他3种估测模型。该结果可为蜜柚叶片钙素含量快速诊断提供新方法以供参考。

关键词: 光谱指数, 估测模型, 高光谱, 蜜柚, 钙素

Abstract:

【Objective】By constructing the estimation model for calcium (Ca) content in honey pomelo leaves based on spectral analysis, it could provide a theoretical basis for monitoring and rapid non-destructive diagnosis of Ca content in honey pomelo leaves.【Method】The original spectral and first-order derivative spectral characteristic bands and spectral characteristic indices (difference spectral index (DSI), ratio spectral index (RSI), and normalized difference spectral index (NDSI)) were analyzed and extracted. Single variable estimation model, partial least squares estimation model (PLS), backpropagation neural network estimation model (BPNN), random forest estimation model (RF), and support vector machine estimation model (SVM) for honey pomelo leaf calcium content were established, and the optimal spectral estimation model for honey pomelo leaf calcium content was evaluated and verified. 【Result】There was a significant multi band correlation between the original spectrum and first-order derivative spectrum of pomelo leaves and calcium content. Based on the correlation coefficients of the original spectrum and first-order derivative spectrum, the maximum wavelengths were 553, 714 nm and 528, 699, 602 nm, respectively. The spectral indices with significant correlation between the original spectrum, first-order derivative of pomelo leaves and calcium content were DSI790,1040, RSI910,990, NDSI900,990 and NDSI′350,580, DSI′560,570, RSI′350,580. The polynomial estimation model constructed with spectral indices such as RSI910,990, NDSI900,990, NDSI′350,580, DSI790,1040, DSI′560,570, RSI′350,580, DSI′528,602 as independent variables had relatively high determination coefficient R2 (R2>0.60). A hyperspectral estimation model for calcium content in honey pomelo leaves was established using the above four machine learning methods. The R2 of PLS, BPNN, RF and SVM estimation models were 0.79, 0.82, 0.85 and 0.84, respectively, and the root mean square errors (RMSE) were 4.33, 4.11, 3.81 and 3.93, respectively; the R2 of the validation models were 0.77, 0.80, 0.87 and 0.83, respectively, and the RMSE were 4.50, 4.28, 3.67 and 3.90, respectively. The order of estimating the accuracy of the model was RF>SVM>BPNN>PLS.【Conclusion】The accuracy comparison analysis of four models for calcium content in honey pomelo leaves showed that the RF estimation model had better predictive performance than the other three estimation models. This result could provide a new method for rapid diagnosis of calcium content in honey pomelo leaves for reference.

Key words: spectral index, estimation model, hyperspectrum, honey pomelo, Calcium