Accurately monitoring alfalfa nutritional quality is essential for optimal pasture management. Yet, current UAV remote sensing methods rely on single-temporal imagery and empirical indices, limiting their ability to handle multi-stage growth dynamics, canopy spectral saturation, and canopy-to-whole-plant scale differences. Furthermore, small sample sizes often cause purely data-driven models to overfit correlations, yielding biologically unrealistic results. Overcoming these challenges, we designed a comprehensive quality estimation framework using 127 alfalfa core germplasms, combining high-dimensional spectral mining, a physics-informed network, and a 3D allometric transfer operator. After screening 14,960 spectral operators across original and log-transformed spaces, we applied a dual dimensionality reduction strategy to isolate optimal features. Four-band dual-difference structures proved highly sensitive to fiber components (ADF/NDF, |r| = 0.896), while logarithmic decoupling operators accurately isolated protein and nitrogen signals (CP/N, |r| = 0.868). We then engineered a Physics-Informed Sparse Shallow Network (PI-SSN). By leveraging temporal attention decoupling, it adaptively assigns growth-stage weights to different components and uses carbon-nitrogen metabolic constraints to maintain biological accuracy during multi-task retrieval. Multi-stage temporal data significantly boosted accuracy over single-period spectra. PI-SSN delivered exceptional test set coefficients of determination ( R2 ) of 0.812-0.848 and RPDs >2.0 for N, CP, ADF, and NDF, easily outperforming standard baselines. To bridge the canopy-only observation gap, we introduced a 3D allometric transfer operator that incorporates canopy coverage and plant height. This effectively corrected vertical stem-leaf observation biases, enhancing Relative Feed Value (RFV) predictions. Ultimately, this approach offers a powerful new framework for high-throughput forage phenotyping.
Why it matches plant phenotyping methodsUAVリモートセンシングと物理制約ネットワーク、3Dアロメトリック演算子を統合し、アルファルファの栄養品質を推定する手法を開発・検証しており、植物表現型取得が中心である。
abstractwe designed a comprehensive quality estimation framework using 127 alfalfa core germplasms, combining high-dimensional spectral mining, a physics-informed network, and a 3D allometric transfer operator.
Reproduction assets foundThe paper's authors publicly release the pre-trained PI-SSN model weights, inference code, and usage instructions on GitHub. The raw spectral and ground-truth quality datasets are not public and are available only on request, so they do not qualify as public assets.Code · publiceptualization, Resources, Supervision, Writing-review & editing. Dongyan Zhang: Conceptualization, Funding acquisition, Project Administration, Supervision, Writing-original draft, Writing-review & editing.
Data and code availability
The pre-trained model weights, inference code, and usage instructions are publicly available at https://github.com/AeroPheno/PI-SSN.git . The raw spectral data and ground-truth quality data used in this study are not publicly available due to ongoing collaborative projects, but are available from the corresponding author on reasonable request.
Funding
This work was supported by the 2023 Hohhot to introduce high-level innovative and entrepreneurial talents (teamOpen asset ↗AeroPheno/PI-SSNlines:243-301Code / dataset availability confirmedOpenAlex · Crossref · checked 14 Sept 2026
Alfalfa is a deep-rooted perennial forage crop with diverse drought-tolerant traits. This study evaluated 250 alfalfa half-sib populations over three growing seasons (2021–2023) under irrigated and rainfed conditions in the Mediterranean drought-prone region of Central Chile (Cauquenes), aiming to identify high-yielding, drought-tolerant populations using remote sensing. Specifically, we assessed RGB-derived indices and canopy temperature difference (CTD; Tc − Ta) as proxies for forage yield (FY). The results showed considerable variation in FY across populations. Under rainfed conditions, winter FY ranged from 1.4 to 6.1 Mg ha−1 and total FY from 3.7 to 14.7 Mg ha−1. Under irrigation, winter FY reached up to 8.2 Mg ha−1 and total FY up to 25.1 Mg ha−1. The AlfaL4-5 (SARDI7), AlfaL57-7 (WL903), and AlfaL62-9 (Baldrich350) populations consistently produced the highest yields across regimes. RGB indices such as hue, saturation, b*, v*, GA, and GGA positively correlated with FY, while intensity, lightness, a*, and u* correlated negatively. CTD showed a significant negative correlation with FY across all seasons and water regimes. These findings highlight the potential of RGB imaging and CTD as effective, high-throughput field phenotyping tools for selecting drought-resilient alfalfa genotypes in Mediterranean environments.
Why it matches plant phenotyping methodsRGB画像指標と冠層温度差を用いた高スループット表現型解析を、アルファルファ集団の収量・干ばつ耐性選抜に実質的に適用しており、表現型取得手法が中心的である。
titleSelecting High Forage-Yielding Alfalfa Populations in a Mediterranean Drought-Prone Environment Using High-Throughput Phenotyping
Reproduction assets found保存済みの本文根拠を更新済みルールで再検証し、公開資産1件を確認しました。Code · publicthe Mosaic tool software and Cereal-Scanner plugin,
developed by Shawn Kefauver from the University of Barcelona, were utilized for further
analysis (available at https://gitlab.com/sckefauver/cerealscanner (accessed on 6 March
2025)).Open asset ↗gitlab.com/sckefauver/cerealscannerpdf-page:7 lines:1-55Code / dataset availability confirmedEurope PMC · checked 7 Sept 2026
Multispectral imaging by unoccupied aerial vehicles provides a nondestructive, high-throughput approach to measure biomass accumulation over successive alfalfa (Medicago sativa L. subsp. sativa) harvests. Information from estimated growth curves can be used to infer harvest biomass and to gain insights into the relationship between growth dynamics and forage biomass stability across cuttings and years. In this study, multispectral imaging and several common vegetation indices were used to estimate genetic parameters and model growth of alfalfa cultivars to determine the longitudinal relationship between vegetation indices and forage biomass. Results showed moderate heritability for vegetation indices, with median plot level heritability ranging from 0.11 to 0.64, across multiple cuttings in three trials planted in Ithaca, NY, and Las Cruces, NM. Genetic correlations between the normalized difference vegetation index and forage biomass were moderate to high across trials, cuttings, and the timing of multispectral image capture. To evaluate the relationship between growth parameters and forage biomass stability across cuttings and environmental conditions, random regression modeling approaches were used to estimate the growth parameters of cultivars for each cutting and the variance in growth was compared to the variance in genetic estimates of forage biomass yield across cuttings. These analyses revealed high correspondence between stability in growth parameters and stability of forage yield. The results of this study indicate that vegetation indices are effective at modeling genetic components of biomass accumulation, presenting opportunities for more efficient screening of cultivars and new longitudinal modeling approaches that can provide insights into temporal factors influencing cultivar stability.
Why it matches plant phenotyping methodsUAVマルチスペクトル画像と植生指数を用いてアルファルファのバイオマス蓄積を推定・モデル化し、遺伝パラメータや生育安定性を評価することが研究の中心であるため。
abstractMultispectral imaging by unoccupied aerial vehicles provides a nondestructive, high-throughput approach to measure biomass accumulation over successive alfalfa (Medicago sativa L. subsp. sativa) harvests.
Reproduction assets foundThe paper's authors publicly deposited the R analysis code and input data for the random regression growth-curve modeling and stability analysis in a GitHub repository, explicitly stated in the Data availability section. Phenotype/imagery data themselves are only available upon request (request_only), and Pix4D is a第三方Code · publicAll data is available upon request. R Code and input data are available in the github: https://github.com/rthapa1/FFAR_RandomRegressionModel_growthcurve_modelling_stabilityanalysis_alfalfa .Open asset ↗rthapa1/FFAR_RandomRegressionModel_growthcurve_modelling_stabilityanalysis_alfalfalines:145-180Code / dataset availability confirmedEurope PMC · checked 7 Sept 2026
Alfalfa / lucerneRootClassificationRoot system architecture
Background: Root system architecture (RSA) is of growing interest in implementing plant improvements with belowground root traits. Modern computing technology applied to images offers new pathways forward to plant trait improvements and selection through RSA analysis (using images to discern/classify root types and traits). However, a major stumbling block to image-based RSA phenotyping is image label noise, which reduces the accuracies of models that take images as direct inputs. To address the label noise problem, this study utilized an artificial intelligence model capable of classifying the RSA of alfalfa ( Medicago sativa L.) directly from images and coupled it with downstream label improvement methods. Images were compared with different model outputs with manual root classifications, and confident machine learning (CL) and reactive machine learning (RL) methods were tested to minimize the effects of subjective labeling to improve labeling and prediction accuracies. Results: The CL algorithm modestly improved the Random Forest model's overall prediction accuracy of the Minnesota dataset (1%) while larger gains in accuracy were observed with the ResNet-18 model results. The ResNet-18 cross-population prediction accuracy was improved (~8% to 13%) with CL compared to the original/preprocessed datasets. Training and testing data combinations with the highest accuracies (86%) resulted from the CL- and/or RL-corrected datasets for predicting taproot RSAs. Similarly, the highest accuracies achieved for the intermediate RSA class resulted from corrected data combinations. The highest overall accuracy (~75%) using the ResNet-18 model involved CL on a pooled dataset containing images from both sample locations. Conclusions: ResNet-18 DNN prediction accuracies of alfalfa RSA image labels are increased when CL and RL are employed. By increasing the dataset to reduce overfitting while concurrently finding and correcting image label errors, it is demonstrated here that accuracy increases by as much as ~11% to 13% can be achieved with semi-automated, computer-assisted preprocessing and data cleaning (CL/RL).
Why it matches plant phenotyping methodsアルファルファ根系構造を画像から分類するResNet-18と、ラベルノイズを補正する機械学習手法を開発・評価しており、植物形質取得ワークフローが中心である。
abstracta major stumbling block to image-based RSA phenotyping is image label noise
Reproduction assets foundThe paper's Data Availability statement deposits two paper-specific public assets on Zenodo: the Minnesota alfalfa root crown images (with tags removed and RootPainter-segmented images) and the Oklahoma root crown images together with the R statistical analysis code generated in this study. Both are directly usable, soDataset · publicThe original images (dataset 1 from USDA-ARS at St Paul, MN) with tags removed and segmented images from RootPainter for data analysis are available on Zenodo ( https://doi.org/10.5281/zenodo.5879778 ).Open asset ↗Zenodo · 10.5281/zenodo.5879778lines:341-365Dataset · publicDataset 2 from Oklahoma: Root crown images and R statistical analysis code generated from this study are available on Zenodo ( https://doi.org/10.5281/zenodo.2172832 ).Open asset ↗Zenodo · 10.5281/zenodo.2172832lines:341-365Code / dataset availability confirmedEurope PMC · checked 8 Sept 2026
Alfalfa / lucerneField / plotRootClassificationRoot system architecture
Active breeding programs specifically for root system architecture (RSA) phenotypes remain rare; however, breeding for branch and taproot types in the perennial crop alfalfa is ongoing. Phenotyping in this and other crops for active RSA breeding has mostly used visual scoring of specific traits or subjective classification into different root types. While image-based methods have been developed, translation to applied breeding is limited. This research is aimed at developing and comparing image-based RSA phenotyping methods using machine and deep learning algorithms for objective classification of 617 root images from mature alfalfa plants collected from the field to support the ongoing breeding efforts. Our results show that unsupervised machine learning tends to incorrectly classify roots into a normal distribution with most lines predicted as the intermediate root type. Encouragingly, random forest and TensorFlow-based neural networks can classify the root types into branch-type, taproot-type, and an intermediate taproot-branch type with 86% accuracy. With image augmentation, the prediction accuracy was improved to 97%. Coupling the predicted root type with its prediction probability will give breeders a confidence level for better decisions to advance the best and exclude the worst lines from their breeding program. This machine and deep learning approach enables accurate classification of the RSA phenotypes for genomic breeding of climate-resilient alfalfa.
Why it matches plant phenotyping methodsアルファルファ根系構造を対象に、画像増強と機械学習・深層学習による表現型分類手法を開発・比較しており、フェノタイピング手法が研究の中心である。
abstractThis research is aimed at developing and comparing image-based RSA phenotyping methods using machine and deep learning algorithms
Reproduction assets foundThe paper's root images (originals with tags removed and RootPainter segmentations) used for the alfalfa RSA phenotyping/ML analysis are publicly deposited on Zenodo (doi: 10.5281/zenodo.5879778), as stated in the Data Availability section. No allowed URL in the supplied list matches this deposit, so no URL is providedDataset · publicThe original images with tags removed and segmented images from RootPainter for data analysis are available on Zenodo doi: 10.5281/zenodo.5879778 [ 85 ].Zenodo · 10.5281/zenodo.5879778lines:627-653Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
Seed vigor is an important index to evaluate seed quality in plant species. How to evaluate seed vigor quickly and accurately has always been a serious problem in the seed research field. As a new physical testing method, multispectral technology has many advantages such as high sensitivity and accuracy, nondestructive and rapid application having advantageous prospects in seed quality evaluation. In this study, the morphological and spectral information of 19 wavelengths (365, 405, 430, 450, 470, 490, 515, 540, 570, 590, 630, 645, 660, 690, 780, 850, 880, 940, 970 nm) of alfalfa seeds with different level of maturity and different harvest periods (years), representing different vigor levels and age of seed, were collected by using multispectral imaging. Five multivariate analysis methods including principal component analysis (PCA), linear discriminant analysis (LDA), support vector machine (SVM), random forest (RF) and normalized canonical discriminant analysis (nCDA) were used to distinguish and predict their vigor. The results showed that LDA model had the best effect, with an average accuracy of 92.9% for seed samples of different maturity and 97.8% for seed samples of different harvest years, and the average sensitivity, specificity and precision of LDA model could reach more than 90%. The average accuracy of nCDA in identifying dead seeds with no vigor reached 93.3%. In identifying the seeds with high vigor and predicting the germination percentage of alfalfa seeds, it could reach 95.7%. In summary, the use of Multispectral Imaging and multivariate analysis in this experiment can accurately evaluate and predict the seed vigor, seed viability and seed germination percentages of alfalfa, providing important technical methods and ideas for rapid non-destructive testing of seed quality.
Why it matches plant phenotyping methodsマルチスペクトル画像と多変量解析により、アルファルファ種子の活力・生存性・発芽率を非破壊推定する手法が研究の中心であるため。
abstractAs a new physical testing method, multispectral technology has many advantages such as high sensitivity and accuracy, nondestructive and rapid application having advantageous prospects in seed quality evaluation.
Reproduction assets foundThe authors provide a public Google Drive supplement containing the paper's own multispectral imaging data: mean reflectance at 19 wavelengths for all seeds (Table S1), morphological feature data for all seeds (Table S2), and multispectral images of the alfalfa seed samples (Figures S1–S6). These directly reproduce theDataset · publicThe following are available online at https://drive.google.com/file/d/13CXchEm81qnbIZCXLqdvupDPib7BS8FM/view?usp=sharing , Table S1: Mean reflectance of 19 wavelengths in all seeds, Table S2: Data of morphological feature in all seeds. Figure S1: Multispectral image of seeds harvested in 2004. Figure S2: Multispectral image of seeds harvested in 2008. Figure S3: Multispectral image of seeds harvested in 2019. Figure S4: Multispectral image of seeOpen asset ↗lines:79-239Code / dataset availability confirmedbioRxiv · Europe PMC · checked 15 Sept 2026
Root system architecture (RSA) is critical for plant growth, which is influenced by several edaphic, environmental, genetic and biotic factors including beneficial and pathogenic microbes. Studying root architecture and the dynamic changes that occur during a plants lifespan, especially for perennial crops growing over multiple growing seasons, is still a challenge because of the nature of their growing environment in soil. We describe the utility of an imaging platform called RhizoVision Crown to study RSA of alfalfa, a perennial forage crop affected by Phymatotrichopsis Root Rot (PRR) disease. Phymatotrichopsis omnivora is the causal agent of PRR disease that reduces alfalfa stand longevity. During the lifetime of the stand, PRR disease rings enlarge and the field can be categorized into three zones based upon plant status: asymptomatic, disease front and survivor. To study root architectural changes associated with PRR, a four-year old 25.6-hectare alfalfa stand infested with PRR was selected at the Red River Farm, Burneyville, OK during October 2017. Line transect sampling was conducted from four actively growing PRR disease rings. At each disease ring, six line transects were positioned spanning 15 m on either side of the disease front with one alfalfa root sampled at every 3 m interval. Each alfalfa root was imaged with the RhizoVision Crown platform using a backlight and a high-resolution monochrome CMOS camera enabling preservation of the natural root architectural integrity. The platforms image analysis software, RhizoVision Analyzer, automatically segmented images, skeletonized, and extracted a suite of features. Data indicated that the survivor plants compensated for damage or loss to the taproot through the development of more lateral and crown roots, and that a suite of multivariate features could be used to automatically classify roots as from survivor or asymptomatic zones. Root growth is a dynamic process adapting to ever changing interactions among various phytobiome components, by utilizing a low-cost, efficient and high-throughput Rhizo-Vision Crown platform we showed quantification of these changes occurring in a mature perennial forage crop.
Why it matches plant phenotyping methodsRhizoVision CrownとRhizoVision Analyzerによる根系形態の画像取得・自動解析が研究の中心であり、根系構造特徴の抽出と分類を実施しているため、植物フェノタイピング手法として含める。
abstractWe describe the utility of an imaging platform called RhizoVision Crown to study RSA of alfalfa
Reproduction assets foundThe paper's Data Availability section explicitly deposits the root crown images and R statistical analysis code on Zenodo (doi 10.5281/zenodo.2172832), a paper-specific public asset containing the phenotyping images and analysis code.Dataset · publicical analysis code generated from this study are available on
382
Zenodo.
383
York, Larry M., Young, Carolyn A., Mattupalli, Chakradhar, & Seethepalli, Anand. (2018). Images
384
and statistical analysis of alfalfa root crowns from inside and outside disease rings caused by
385
cotton root rot (Version 1.0.0) [Data set]. Zenodo. http://doi.org/10.5281/zenodo.2172832
386
387
ACKNOWLEDGEMENTS. We thank the Noble Research Institute, LLC for funding this project.
388
389
LITERATURE CITED.
390
Arias, M. M. D., Leandro, L. F., and Munkvold, G. P. 2013. Aggressiveness of Fusarium species and
391
impact of root infection on growth and yield of soybeans. Phytopathology 103:822-832.
392
Arif, M., FlOpen asset ↗Zenodo · 10.5281/zenodo.2172832pdf-raw-page:18 lines:1-49