Parida Pradosh Kumar, Eagan Somasundaram, Ramanujam Krishnan, Sengodan Radhamani, Uthandi Sivakumar, Ettiyagounder Parameswari, Rajagounder Raja
Department of Agronomy, Tamil Nadu Agricultural University, Coimbatore, 641003, Tamil Nadu, India.
Directorate of Agribusiness Development (DABD), Tamil Nadu Agricultural University, Coimbatore, 641003, Tamil Nadu, India.
Heliyon. 2024 Jul 4;10(13):e34117. doi: 10.1016/j.heliyon.2024.e34117. eCollection 2024 Jul 15.
The fraction of absorbed photosynthetically active radiation (FAPAR) and the photosynthesis rate (Pn) of maize canopies were identified as essential photosynthetic parameters for accurately estimating vegetation growth and productivity using multispectral vegetation indices (VIs). Despite their importance, few studies have compared the effectiveness of multispectral imagery and various machine learning techniques in estimating these photosynthetic traits under high vegetation coverage. In this study, seventeen multispectral VIs and four machine learning (ML) algorithms were utilized to determine the most suitable model for estimating maize FAPAR and Pn during the and seasons at Tamil Nadu Agricultural University, Coimbatore, India. Results demonstrate that indices such as OSAVI, SAVI, EVI-2, and MSAVI-2 during the and MNDVIRE and MSRRE during the season outperformed others in estimating FAPAR and Pn values. Among the four ML methods of random forest (RF), extreme gradient boosting (XGBoost), support vector regression (SVR), and multiple linear regression (MLR) considered, RF consistently showed the most effective fitting effect and XGBoost demonstrated the least fitting accuracy for FAPAR and Pn estimation. However, SVR with R = 0.873 and RMSE = 0.045 during the and MLR with R = 0.838 and RMSE = 0.053 during the season demonstrated higher fitting accuracy, particularly notable for FAPAR prediction. Similarly, in the prediction of Pn, MLR showed higher fitting accuracy with R = 0.741 and RMSE = 2.531 during the and R = 0.955 and RMSE = 1.070 during the season. This study demonstrated the potential of combining UAV-derived VIs with ML to develop accurate FAPAR and Pn prediction models, overcoming VI saturation in dense vegetation. It underscores the importance of optimizing these models to improve the accuracy of maize vegetation assessments during various growing seasons.
吸收的光合有效辐射比例（FAPAR）和玉米冠层的光合速率（Pn）被确定为利用多光谱植被指数（VIs）准确估算植被生长和生产力的关键光合参数。尽管它们很重要，但很少有研究比较多光谱图像和各种机器学习技术在高植被覆盖下估算这些光合特征的有效性。在本研究中，利用17种多光谱植被指数和4种机器学习（ML）算法，在印度哥印拜陀的泰米尔纳德邦农业大学，确定了估算玉米FAPAR和Pn在两个季节期间最合适的模型。结果表明，在第一季中，如OSAVI、SAVI、EVI - 2和MSAVI - 2等指数，以及在第二季中MNDVIRE和MSRRE等指数，在估算FAPAR和Pn值方面优于其他指数。在所考虑的随机森林（RF）、极端梯度提升（XGBoost）、支持向量回归（SVR）和多元线性回归（MLR）这四种ML方法中，RF始终显示出最有效的拟合效果，而XGBoost在FAPAR和Pn估算方面显示出最低的拟合精度。然而，第一季中R = 0.873且RMSE = 0.045的SVR和第二季中R = 0.838且RMSE = 0.053的MLR显示出更高的拟合精度，特别是在FAPAR预测方面尤为显著。同样，在Pn预测中，MLR在第一季中R = 0.741且RMSE = 2.531，在第二季中R = 0.955且RMSE = 1.070时显示出更高的拟合精度。本研究证明了将无人机获取的植被指数与机器学习相结合，以开发准确的FAPAR和Pn预测模型的潜力，克服了茂密植被中植被指数的饱和问题。它强调了优化这些模型对于提高不同生长季节玉米植被评估准确性的重要性。
