Frontiers in Plant Science (Jun 2024)
Estimation of soybean yield based on high-throughput phenotyping and machine learning
- Xiuni Li,
- Xiuni Li,
- Xiuni Li,
- Menggen Chen,
- Menggen Chen,
- Menggen Chen,
- Shuyuan He,
- Shuyuan He,
- Shuyuan He,
- Xiangyao Xu,
- Xiangyao Xu,
- Xiangyao Xu,
- Lingxiao He,
- Lingxiao He,
- Lingxiao He,
- Li Wang,
- Li Wang,
- Li Wang,
- Yang Gao,
- Yang Gao,
- Yang Gao,
- Fenda Tang,
- Fenda Tang,
- Fenda Tang,
- Tao Gong,
- Tao Gong,
- Tao Gong,
- Wenyan Wang,
- Wenyan Wang,
- Wenyan Wang,
- Mei Xu,
- Mei Xu,
- Mei Xu,
- Chunyan Liu,
- Chunyan Liu,
- Chunyan Liu,
- Liang Yu,
- Liang Yu,
- Liang Yu,
- Weiguo Liu,
- Weiguo Liu,
- Weiguo Liu,
- Wenyu Yang,
- Wenyu Yang,
- Wenyu Yang
Affiliations
- Xiuni Li
- College of Agronomy, Sichuan Agricultural University, Chengdu, China
- Xiuni Li
- Sichuan Engineering Research Center for Crop Strip Intercropping System, Chengdu, China
- Xiuni Li
- Key Laboratory of Crop Ecophysiology and Farming System in Southwest, Ministry of Agriculture, Chengdu, China
- Menggen Chen
- College of Agronomy, Sichuan Agricultural University, Chengdu, China
- Menggen Chen
- Sichuan Engineering Research Center for Crop Strip Intercropping System, Chengdu, China
- Menggen Chen
- Key Laboratory of Crop Ecophysiology and Farming System in Southwest, Ministry of Agriculture, Chengdu, China
- Shuyuan He
- College of Agronomy, Sichuan Agricultural University, Chengdu, China
- Shuyuan He
- Sichuan Engineering Research Center for Crop Strip Intercropping System, Chengdu, China
- Shuyuan He
- Key Laboratory of Crop Ecophysiology and Farming System in Southwest, Ministry of Agriculture, Chengdu, China
- Xiangyao Xu
- College of Agronomy, Sichuan Agricultural University, Chengdu, China
- Xiangyao Xu
- Sichuan Engineering Research Center for Crop Strip Intercropping System, Chengdu, China
- Xiangyao Xu
- Key Laboratory of Crop Ecophysiology and Farming System in Southwest, Ministry of Agriculture, Chengdu, China
- Lingxiao He
- College of Agronomy, Sichuan Agricultural University, Chengdu, China
- Lingxiao He
- Sichuan Engineering Research Center for Crop Strip Intercropping System, Chengdu, China
- Lingxiao He
- Key Laboratory of Crop Ecophysiology and Farming System in Southwest, Ministry of Agriculture, Chengdu, China
- Li Wang
- College of Agronomy, Sichuan Agricultural University, Chengdu, China
- Li Wang
- Sichuan Engineering Research Center for Crop Strip Intercropping System, Chengdu, China
- Li Wang
- Key Laboratory of Crop Ecophysiology and Farming System in Southwest, Ministry of Agriculture, Chengdu, China
- Yang Gao
- College of Agronomy, Sichuan Agricultural University, Chengdu, China
- Yang Gao
- Sichuan Engineering Research Center for Crop Strip Intercropping System, Chengdu, China
- Yang Gao
- Key Laboratory of Crop Ecophysiology and Farming System in Southwest, Ministry of Agriculture, Chengdu, China
- Fenda Tang
- College of Agronomy, Sichuan Agricultural University, Chengdu, China
- Fenda Tang
- Sichuan Engineering Research Center for Crop Strip Intercropping System, Chengdu, China
- Fenda Tang
- Key Laboratory of Crop Ecophysiology and Farming System in Southwest, Ministry of Agriculture, Chengdu, China
- Tao Gong
- College of Agronomy, Sichuan Agricultural University, Chengdu, China
- Tao Gong
- Sichuan Engineering Research Center for Crop Strip Intercropping System, Chengdu, China
- Tao Gong
- Key Laboratory of Crop Ecophysiology and Farming System in Southwest, Ministry of Agriculture, Chengdu, China
- Wenyan Wang
- College of Agronomy, Sichuan Agricultural University, Chengdu, China
- Wenyan Wang
- Sichuan Engineering Research Center for Crop Strip Intercropping System, Chengdu, China
- Wenyan Wang
- Key Laboratory of Crop Ecophysiology and Farming System in Southwest, Ministry of Agriculture, Chengdu, China
- Mei Xu
- College of Agronomy, Sichuan Agricultural University, Chengdu, China
- Mei Xu
- Sichuan Engineering Research Center for Crop Strip Intercropping System, Chengdu, China
- Mei Xu
- Key Laboratory of Crop Ecophysiology and Farming System in Southwest, Ministry of Agriculture, Chengdu, China
- Chunyan Liu
- College of Agronomy, Sichuan Agricultural University, Chengdu, China
- Chunyan Liu
- Sichuan Engineering Research Center for Crop Strip Intercropping System, Chengdu, China
- Chunyan Liu
- Key Laboratory of Crop Ecophysiology and Farming System in Southwest, Ministry of Agriculture, Chengdu, China
- Liang Yu
- College of Agronomy, Sichuan Agricultural University, Chengdu, China
- Liang Yu
- Sichuan Engineering Research Center for Crop Strip Intercropping System, Chengdu, China
- Liang Yu
- Key Laboratory of Crop Ecophysiology and Farming System in Southwest, Ministry of Agriculture, Chengdu, China
- Weiguo Liu
- College of Agronomy, Sichuan Agricultural University, Chengdu, China
- Weiguo Liu
- Sichuan Engineering Research Center for Crop Strip Intercropping System, Chengdu, China
- Weiguo Liu
- Key Laboratory of Crop Ecophysiology and Farming System in Southwest, Ministry of Agriculture, Chengdu, China
- Wenyu Yang
- College of Agronomy, Sichuan Agricultural University, Chengdu, China
- Wenyu Yang
- Sichuan Engineering Research Center for Crop Strip Intercropping System, Chengdu, China
- Wenyu Yang
- Key Laboratory of Crop Ecophysiology and Farming System in Southwest, Ministry of Agriculture, Chengdu, China
- DOI
- https://doi.org/10.3389/fpls.2024.1395760
- Journal volume & issue
-
Vol. 15
Abstract
IntroductionSoybeans are an important crop used for food, oil, and feed. However, China’s soybean self-sufficiency is highly inadequate, with an annual import volume exceeding 80%. RGB cameras serve as powerful tools for estimating crop yield, and machine learning is a practical method based on various features, providing improved yield predictions. However, selecting different input parameters and models, specifically optimal features and model effects, significantly influences soybean yield prediction.MethodsThis study used an RGB camera to capture soybean canopy images from both the side and top perspectives during the R6 stage (pod filling stage) for 240 soybean varieties (a natural population formed by four provinces in China: Sichuan, Yunnan, Chongqing, and Guizhou). From these images, the morphological, color, and textural features of the soybeans were extracted. Subsequently, feature selection was performed on the image parameters using a Pearson correlation coefficient threshold ≥0.5. Five machine learning methods, namely, CatBoost, LightGBM, RF, GBDT, and MLP, were employed to establish soybean yield estimation models based on the individual and combined image parameters from the two perspectives extracted from RGB images.Results(1) GBDT is the optimal model for predicting soybean yield, with a test set R2 value of 0.82, an RMSE of 1.99 g/plant, and an MAE of 3.12%. (2) The fusion of multiangle and multitype indicators is conducive to improving soybean yield prediction accuracy.ConclusionTherefore, this combination of parameters extracted from RGB images via machine learning has great potential for estimating soybean yield, providing a theoretical basis and technical support for accelerating the soybean breeding process.
Keywords