Plant Phenomics最新文献

With the increasing global demand for food, breeding soybean varieties resistant to dense planting is crucial for achieving high and stable yields. Traditional phenotyping methods are limited by insufficient temporal resolution and challenges in dynamic modeling continuity, making it difficult to elucidate the intrinsic relationship between canopy development rate and yield stability. Moreover, existing machine learning models often neglect temporal dependencies in time series predictions, leading to insufficient biological interpretability. This study proposes an innovative approach integrating spatiotemporal deep learning and dynamic modeling to quantify the dynamic changes in canopy parameters using UAV high-throughput phenotyping technology, revealing the key regulatory mechanisms of traits associated with resistance to dense planting. Based on a two-year field experiment (2022-2023) in northeast China (Qiqihaer, black soil region), this study set high (50w plants/ha) and low density (30w plants/ha) treatments across 208 soybean varieties, combined with multispectral UAV imagery (15-18 times per season) and ground-truth data, to develop a time series prediction model for leaf area index (LAI). Comparing the performance of spatiotemporal residual networks (ST-ResNet), long short-term memory networks (LSTM), and traditional random forests (RF), the ST-ResNet model demonstrated significantly superior prediction accuracy (R² ​= ​0.90, RMSE ​= ​0.23 ​m²/m²), effectively capturing the continuous dynamics of canopy growth through its spatiotemporal feature fusion ability. By fitting the time series curves of LAI, canopy cover (CC), and plant height (PH) with P-spline, 15 intermediate traits (e.g., ΔMean_LAI-mid) were extracted. Mixed models and SHAP interpretability analysis showed that ΔMean_LAI-mid was most correlated with the dense planting yield index (ΔYield, r ​= ​0.51). Furthermore, the high-frequency data acquisition and automated analysis framework using UAVs enabled high-throughput phenotypic screening for 208 varieties per year, significantly improving efficiency compared to traditional methods that rely on manual sampling. This study pioneers the integration of spatiotemporal deep learning with dynamic trait modeling, markedly improving the temporal continuity and stability of LAI estimation compared to traditional single-time-point prediction methods. This advancement allows for more precise quantification of canopy development rates across various growth stages, enabling a systematic analysis of how these dynamic patterns influence resistance to dense planting. By elucidating the dynamic relationship between intermediate traits and yield, this approach offers a high-precision, interpretable phenotypic analysis framework for effectively screening soybean varieties resilient to dense planting.

Digital phenotyping is rapidly advancing, generating increasing amounts of data, particularly in the case of temporal monitoring. We propose an adaptive sampling method that optimizes sampling, thereby reducing costs associated with data production, processing, and storage. The proposed method is based on Bayesian inference, which utilizes previous measurements, historical data, and an expected model. Five Bayesian methods are assessed in this study: Important sampling (IS), Markov chain Monte-Carlo (MCMC), Gaussian process (GP), Extended Kalman filtering (EKF) and Sampling Importance Resampling particle filtering (SIR-PF). We test these five Bayesian sampling methods for the monitoring of germination rate in terms of compression, distortion and computation cost. The best trade-off is found by the MCMC method, which offers a compression rate of 0.2 with very little distortion. GP offers the most unbiased parameter estimation and the capability to adapt to various germination speeds. It also has reasonable computational times.

Weed growth significantly impacts corn yield. With the continuous development of weed control technologies, achieving more effective and precise weed management has become a major challenge in corn production. To achieve precise weed suppression, this study proposes a growth point detection method based on a keypoint pose estimation model capable of effectively detecting various weeds and locating various weed growth points during the 2nd-5th leaf stage of corn development. To address the complex working environment of precision weeding machines in corn fields, including occlusion, dense growth, and variable lighting conditions, we design a dilation-wise residual module (DWRM) for the detector and a separation and enhancement attention module (SEAM) for pose estimation to adapt to these challenges. Furthermore, owing to the limited computational resources in field settings, we introduced the RepViT block (RVB) to achieve model lightweighting. The proposed method was evaluated on the constructed corn field dataset. The experimental results demonstrated that SRD-YOLO achieved an ${mAP}_{kpt}$ of 96.5 ​%, an F1 score of 94 ​%, and an FPS of 169, while reducing the model parameters by 8.7M. SRD-YOLO effectively meets the requirements for growth point localization under challenging conditions, providing robust technical support for real-time and precise weed control in corn fields.

Today, leaf trait estimation remains a labor-intensive process. The effort to obtain ground truth measurements limits how accurately this task can be performed automatically. Traditionally, plant scientists manually measure the traits of harvested leaves and associate them with sensor data, which is key for training machine learning approaches and to automate the processes. In this paper, we propose a neural network-based method to generate synthetic 3D point clouds of leaves with their associated traits to support approaches for phenotyping. We use real-world leaf point clouds to learn how to generate realistic leaves from a leaf skeleton, which is automatically extracted. We use the generated leaves to fine-tune different leaf trait estimation methods. We evaluate our generated data using different trait estimation methods and compare the results to using real-world data or other synthetic datasets from agricultural simulation software. Experiments show that our approach generates leaf point clouds with high similarity to real-world leaves. Tuning trait estimation methods on our generated data improves their performance in the estimation of real-world leaves' traits, making our data crucial for developing and testing data-driven trait estimation methods. Accurate trait estimation is key to understanding crop growth, productivity, and pest resistance, as leaf size directly influences photosynthesis, yield potential, and vulnerability to insects and fungal growth.

Forest phenotypic responses are significantly influenced by extreme climate conditions, particularly canopy structure and photosynthetic traits. However, the underlying mechanisms driving these responses, especially in conifer species, remain poorly understood. This study employs advanced phenotyping technologies, combining three-dimensional (3D) canopy reconstruction with high-resolution physiological trait analysis, quantifying changes in key physiological traits that light interception, gas exchange parameters stomatal conductance, and chlorophyll content. Developing 3D reconstruction algorithms tailored to conifer canopies is essential for simulating forest ecosystem responses under varying canopy densities. We investigate the following questions: (1) How does thinning affect canopy light penetration and photosynthetic efficiency? Thinning significantly increased light penetration from 15 ​% (CK) to 22 ​%, enhancing photosynthetic efficiency, resulting in an 18 ​% increase in carbon absorption under drought conditions. (2) How does reduced-rainfall affect photosynthetically active radiation (PAR) and stomatal conductance? Reduced-rainfall caused a 12 ​% decrease in PAR, a 20 ​% reduction in stomatal conductance, and an 8 ​% decrease in chlorophyll content. (3) What are the synergistic effects of thinning and reduced-rainfall in carbon absorption? Thinning under reduced-rainfall increased carbon absorption by 25 ​%. This study reveals a significant correlation between chlorophyll content, leaf nitrogen content, and canopy structural dynamics under drought and elevated temperature conditions, offering new insights into the adaptive mechanisms plants employ to adjust their photosynthetic processes. In conclusion, the development of 3D reconstruction algorithms tailored for conifer canopies, in regulating photosynthetic traits, is crucial for improving forest adaptation, contributing to functional trait-based forest management and ecosystem modeling.

Reliable and automated three-dimensional segmentation of plant organs is essential for extracting phenotypic traits at the organ level. However, existing methods for plant organ segmentation predominantly rely on fully supervised learning, which still necessitates extensive point-by-point annotated datasets and fails to overcome the challenges associated with annotating plant point cloud data. In recent years, self-supervised learning-based point cloud segmentation methods have garnered widespread attention in both industry and academia because of their potential to alleviate the difficulties of point cloud data annotation to some extent. In this study, the paradigm of self-supervised learning is innovatively applied to the field of plant phenotyping through the development of the Plant-MAE, a self-supervised learning-based point cloud segmentation framework. The innovations of the Plant-MAE include a kernel-based point convolution embedding module and a multiangle feature extraction block (MAFEB) based on attention mechanisms. To validate the effectiveness of the model, extensive experiments were conducted on multiple point cloud datasets, which achieved competitive performance, with average precision, recall, F1 score, and IoU values of 92.08 ​%, 88.50 ​%, 89.80 ​%, and 84.03 ​%, respectively. The Plant-MAE outperforms advanced deep learning networks, including PointNet++, point transformer, and Point-M2AE, achieving average improvements of at least 0.53 ​%, 1.36 ​%, 0.88 ​%, and 2.38 ​% in precision, recall, F1 score, and IoU, respectively. Additionally, on the Pheno4D dataset, only half of the training data were necessary for fine-tuning to achieve performance comparable to that of the point transformer and PointNet++. This study provides technical support for the estimation of crop phenotypic parameters, thereby advancing the development of modern smart agriculture.

The aboveground biomass (AGB) of crops is an essential metric for monitoring crop growth, making timely and accurate AGB forecasting critical for effective agricultural management. The introduction of Unmanned Aerial Vehicles (UAVs) and advanced sensor technologies has revolutionized traditional AGB prediction techniques. Currently, machine learning (ML) combined with UAV data are commonly utilized, along with the Vegetation Index Weighted Canopy Volume Model (CVM_VI) for AGB prediction. Nevertheless, there is limited investigation into how these methods perform across different agricultural conditions. This study aims to fill this gap by creating specific methodologies for estimating corn AGB under diverse fertilization and irrigation treatments. We utilized LiDAR, multispectral (MS), thermal infrared (TIR), along with measured AGB and Leaf Area Index (LAI) data from various growth stages to develop a stacking ensemble learning model. This model effectively integrates data from multiple sources, resulting in a strong prediction performance with R² of 0.86, Mean Absolute Error (MAE) of 1.54 ​t/ha, and Root Mean Square Error (RMSE) of 2.06 ​t/ha. Meanwhile, the analysis of the accuracy of CVM_VI revealed its efficacy during the early-stage when corn is short, with its predictive capability diminishing as AGB increases. Consequently, we recommend the CVM_VI for early-stage AGB prediction, which can streamline data collection and computational efforts. In contrast, the ML approach, which benefits from data fusion, is more appropriate for predicting AGB during the mid to late growth stages. This study enhances AGB prediction accuracy and speed, providing critical understanding of regional AGB dynamics and supporting better agricultural decision-making.

Herbicide screening requires a substantial amount of time, effort, and cost, making a new herbicide discovery expensive and time-consuming. Various diagnostic methods have been developed, but most of them are destructive and require significant time and effort to identify herbicide activity. Therefore, this study was conducted to apply spectral image analysis for early and rapid diagnosis of herbicidal activity and modes of action (MOAs). RGB, chlorophyll fluorescence (CF), and infrared (IR) thermal images were acquired after treating herbicides with different MOAs to a model plant, oilseed rape (Brassica napus), and analyzed using MATLAB 2021b to quantify NDI, ExG, F _d /F _m , and plant leaf temperature. NDI, ExG and F _d /F _m decreased, while plant leaf temperature increased after herbicide treatment. Distinctive spectral responses were found depending on the herbicide MOAs. PSII and PPO inhibitors showed rapid responses in IR thermal and CF images within 1 day after herbicide treatment. HPPD inhibitor showed a continuous decrease in F _d /F _m , while EPSPS inhibitor showed gradual changes in all spectral indices. Machine learning by Subspace Discriminant algorithm of spectral indices acquired at 6 ​h enabled the diagnosis of herbicide MOAs with 89.6 ​% accuracy, which gradually increased by adding new spectral indices acquired later time points until 3 DAT, when validation accuracy scored 100 ​%. The indices acquired at 6 ​h, and F _d /F _m and leaf temperature data were shown to contribute to higher accuracies of identifying herbicide MOAs. Overall test accuracy scored 87.5 ​%, verifying the possibility of diagnosing herbicide MOAs based on spectral indices. Therefore, we could conclude that herbicide activity and MOAs can be diagnosed by analyzing spectral images combined with machine learning, suggesting the possibility of high-throughput screening of herbicide MOAs using plant image analysis.

Canopy photosynthetic productivity is crucial for the formation of crop yields. Identifying limiting factors and adjustment targets for canopy photosynthesis in specific climates is important for yield increase. However, conducting relevant quantitative research remains challenging. In this study, two typical regions with distinct climatic characteristics were selected for a two-year trial of greenhouse tomatoes grown in different seasons. A three-dimensional canopy photosynthesis model was developed to quantify the factor contributions to the regional differences in accumulated canopy photosynthesis throughout the entire growing season (ACP), and to predict gains in ACP through three scenarios: leaf photosynthetic modifications (S1), plant layout adjustments (S2), and greenhouse film haze increase (S3). The results indicated that differences in ACP were mainly influenced by light environment (LE), leaf photosynthetic physiology (PP), and LE-PP interaction in spring, and canopy structure (CS), PP, LE, and LE-PP interaction in autumn. The predicted ACP enhancement showed as S1 ​> ​S2 ​> ​S3, with S3 showing a more limited effect. The light quantum efficiency under limiting light ( ${\kappa }_{2\text{LL}}$ ) and maximum electron transport rate ( $J_{\max }$ ) were identified as key biochemical phenotypes for tomato high photosynthetic efficiency breeding in different environments. Additionally, adjusting row spacing under current planting density could further improve ACP. Our conclusions could assist researchers in deepening their understanding of canopy photosynthesis limitations under real production conditions, and provide a theoretical foundation for optimizing greenhouse tomato yield in the context of climate change.

Organ instance segmentation of 3D plant point clouds is a crucial prerequisite for organ-level phenotype estimation. However, most current cloud segmentation methods are usually designed for specific crop, hardly fit for both monocotyledonous and dicotyledonous crops which have significant structural differences. This study therefore proposed a two-stage method with higher generalization ability for single-plant organ instance segmentation based on PointNeXt and Quickshift++. The effectiveness of this method was tested on different types of crops. The dataset includes point clouds of 122 self-acquired sugarcanes, 49 open-accessed maizes, and 77 open-accessed tomatoes. The improved PointNeXt model was trained to implement the semantic segmentation of stems and leaves. The average mOA and mIoU on the test set reaches 96.96 ​% and 87.15 ​%, respectively. The Quickshift++ algorithm was then applied to encode the global spatial structure and local connections of plants for rapid localization and segmentation of leaf instance. Our approach outperformed four SOTA methods, ASIS, JSNet, DFSP, and PSegNet in terms of both quantitative and qualitative segmentation results, achieving average values for mPrec, mRec, mF1, and mIoU of 93.32 ​%, 85.60 ​%, 87.94 ​%, and 81.46 ​%, respectively. The proposed method also yields excellent results for several other plants in their early stages, indicating its generalization ability and applicability for organ instance segmentation for different plants, thus providing a powerful tool for plant phenotypic research.