Quantitative Assessment of Soybean Seedling Uniformity Based on Multi-Feature Fusion
DOI:
https://doi.org/10.54691/n5drkx81Keywords:
Soybean seedling uniformity; multi-feature fusion; XGBoost; feature importance; RGB imagery; precision agriculture.Abstract
Accurate assessment of soybean seedling uniformity is essential for evaluating sowing quality and guiding early-stage field management, yet conventional manual inspection remains subjective and inefficient. This study presents a quantitative uniformity assessment framework based on multi-feature fusion using high-resolution RGB images. Per-plant masks and spatial positions are obtained through instance segmentation of soybean seedling images. Five categories of plot-level uniformity features are then extracted—emergence rate, plant spacing, inter-row angle, canopy coverage, and evenness index—along with their means, standard deviations, and coefficients of variation. An XGBoost classifier is trained to map these features to three uniformity grades, achieving an overall accuracy of 86.7% without cross-level misclassification. Feature importance analysis reveals that the standard deviation of plant spacing, emergence rate, and mean plant spacing emerge as the most influential predictors, providing explicit guidance for diagnosing the causes of poor uniformity. The proposed framework offers an interpretable and quantitative solution for automated seedling uniformity evaluation in soybean breeding programs.
Downloads
References
[1] Siega G A, Filho A C P de M, Sharma P, et al. Influence of planting speed on emergency uniformity and distribution of soybean plants[J]. Cerrado: Agricultural and Biological Research, 2024, 1(2): 21-29.
[2] Shirzadifar A, Maharlooei M, Bajwa S G, et al. Mapping crop stand count and planting uniformity using high resolution imagery in a maize crop[J]. Biosystems Engineering, 2020, 200: 377-390.
[3] Hue Y, Kim J H, Nam Y, et al. Changes in the occurrence patterns of rice fungal diseases due to climate change[J]. Research in Plant Disease, 2025, 31(1): 17-29.
[4] Zhou J, Beche E, Vieira C C, et al. Improve soybean variety selection accuracy using UAV-based high-throughput phenotyping technology[J]. Frontiers in Plant Science, 2022, 12.
[5] Zarei E, Yazdi M, Moradi R, et al. Expert judgment and uncertainty in sociotechnical systems analysis[M]//ZAREI E. Safety causation analysis in sociotechnical systems: advanced models and techniques. Cham: Springer Nature Switzerland, 2024: 487-530.
[6] Zhou J, Cui M, Wu Y, et al. Maize (zea mays L.) stem target region extraction and stem diameter measurement based on an internal gradient algorithm in field conditions[J]. Agronomy, 2023, 13(5): 1185.
[7] Yuan H, Song M, Liu Y, et al. Field phenotyping monitoring systems for high-throughput: a survey of enabling technologies, equipment, and research challenges[J]. Agronomy, 2023, 13(11): 2832.
[8] Zhang Q, Luan R, Wang M, et al. Research progress of spectral imaging techniques in plant phenotype studies[J]. Plants, 2024, 13(21): 3088.
[9] Weyler J, Magistri F, Seitz P, et al. In-field phenotyping based on crop leaf and plant instance segmentation[C]//Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision. 2022: 2725-2734.
[10] Vargo M, Aldrich M, Donahue P, et al. Current diagnostic and quantitative techniques in the field of lymphedema management: a critical review[J]. Medical Oncology, 2024, 41(10): 241.
[11] Chen T, Guestrin C. XGBoost: a scalable tree boosting system[C]//Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York, NY, USA: Association for Computing Machinery, 2016: 785-794.
[12] Demir S, Sahin E K. An investigation of feature selection methods for soil liquefaction prediction based on tree-based ensemble algorithms using AdaBoost, gradient boosting, and XGBoost[J]. Neural Computing and Applications, 2023, 35(4): 3173-3190.
[13] Al Kaddouri H, Blaacha J, Hamdaoui H, et al. Optimization of hyperparameters for SVM classification of citrus diseases using grid search and cross-validation[C]//Proceedings of the 4th International Conference on Electronic Engineering and Renewable Energy Systems—Volume 1. Springer, Singapore, 2025: 489-497.
[14] Alhams A, Abdelhadi A, Badri Y, et al. Enhanced bearing fault diagnosis through trees ensemble method and feature importance analysis[J]. Journal of Vibration Engineering & Technologies, 2024, 12(1): 109-125.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Frontiers in Science and Engineering
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
