Agriculture

SenNet: A dual-branch image semantic segmentation network for wheat senescence evaluation and high-yielding variety screening

Abstract

Wheat is one of the three primary staple crops globally, with the senescence of its leaves having a direct effect on yield. However, conventional senescence evaluation methods are mainly based on visual scoring, which are subjective, time-consuming, and hamper the investigation of mechanisms between senescence process and yield formation. High-throughput image-based plant phenotyping techniques offer a promising approach. However, extracting senescence-related semantic information from images presents challenges, including blurred edge segmentation, inadequate characterization of senescence features, and interference from complex field environments. Therefore, this study proposes a dual-branch image senescence segmentation model (SenNet), which integrates edge priors and local–global attention mechanisms, including local–global hierarchical attention mechanisms, gated convolution, and positional encoding modules. First, a wheat senescence dynamics image dataset (19530 images) was constructed, comprising 509 wheat varieties from a two-year and two-replicate field experiments. Then, the SenNet model achieved senescence image segmentation for various wheat varieties, enabling senescence dynamics analysis and high-yielding variety screening. The results showed that: 1) The mean Intersection over Union (mIoU) of the SenNet model was 95.41 %, which represented a 4.01 % improvement over the average mIoU of seven state-of-the-art models. 2) The contributions of the local–global hierarchical attention mechanism, gated convolution, and positional encoding module to the accuracy improvement of SenNet were 3.15 %, 1.62 %, and 1.03 %, respectively. 3) SenNet can be transferred across years and locations. The mIoU accuracy of the SenNet across locations is 96.01 %. Furthermore, the model trained in 2023 can be transferred to 2022 and 2024, achieving mIoU accuracies of 93.75 % and 93.27 %. 4) High-yielding varieties typically experience a later onset of senescence and faster senescence in later stages. Based on the senescence law, this study further constructed new dynamic traits of senescence (e.g., AreaUnderCurve). Leveraging the random forest-based yield prediction (R^2^ = 0.68) from the dynamic traits, high-yielding varieties were screened with an average precision, recall, F1 score, and accuracy of 81 %, 79 %, 80 %, and 87 %, respectively. This study provides an efficient method for monitoring senescence dynamics and predicting yield, offering new insights into the screening of high-yielding varieties.

Enhancing wheat crop physiology monitoring through spectroscopic analysis of stomatal conductance dynamics

Abstract

Monitoring in-vivo stomatal conductance (gs) dynamics is essential for predicting crop water usage and yield sensitivity in response to climate change. Leaf and canopy spectroscopy offer a non-destructive method for gs monitoring; however, the underlying mechanisms connecting leaf spectra with stomatal anatomical and behavioral traits, and their subsequent impacts on gs, remain underexplored. In this study, we conducted a wheat field trial, collecting comprehensive measurements of stomatal anatomical (i.e., size, density) and behavioral (i. e., opening ratio, pore area) traits by a customized, high-resolution microscope, leaf spectra via a handheld spectroradiometer, and gs via a handheld AP4 Leaf Porometer across various genotypes, nitrogen treatments, growth stages, and diurnal environments. We observed substantial gs variability, with stomatal anatomical and behavioral traits jointly accounting for 79% of this variability. We further examined the relationship between leaf spectra and stomatal traits/conductance using a partial least square regression (PLSR) model and discovered that a single PLSR spectral model accurately predicted the variability of each of these traits and gs across our datasets. Furthermore, we demonstrated a strong correspondence between spectral variations resulting from gs and spectral alternation induced by stomatal anatomical and behavioral traits. By analyzing the diurnal association between spectral and gs variability, we revealed important biophysical mechanisms underlying relationships among spectra, stomatal anatomical and behavioral traits, and gs. Collectively, our findings highlight the potential of leaf spectroscopy in advancing crop physiology monitoring, contributing to enhanced food security and sustainability.

Comparison of Different Machine Learning Algorithms for the Prediction of the Wheat Grain Filling Stage Using RGB Images

Abstract

Grain filling is essential for wheat yield formation, but is very susceptible to environmental stresses, such as high temperatures, especially in the context of global climate change. Grain RGB images include rich color, shape, and texture information, which can explicitly reveal the dynamics of grain filling. However, it is still challenging to further quantitatively predict the days after anthesis (DAA) from grain RGB images to monitor grain development. Results: The WheatGrain dataset revealed dynamic changes in color, shape, and texture traits during grain development. To predict the DAA from RGB images of wheat grains, we tested the performance of traditional machine learning, deep learning, and few-shot learning on this dataset. The results showed that Random Forest (RF) had the best accuracy of the traditional machine learning algorithms, but it was far less accurate than all deep learning algorithms. The precision and recall of the deep learning classification model using Vision Transformer (ViT) were the highest, 99.03% and 99.00%, respectively. In addition, few-shot learning could realize fine-grained image recognition for wheat grains, and it had a higher accuracy and recall rate in the case of 5-shot, which were 96.86% and 96.67%, respectively. Materials and Methods: In this work, we proposed a complete wheat grain dataset, WheatGrain, which covers thousands of wheat grain images from 6 DAA to 39 DAA, which can characterize the complete dynamics of grain development. At the same time, we built different algorithms to predict the DAA, including traditional machine learning, deep learning, and few-shot learning, in this dataset, and evaluated the performance of all models. Conclusions: To obtain wheat grain filling dynamics promptly, this study proposed an RGB dataset for the whole growth period of grain development. In addition, detailed comparisons were conducted between traditional machine learning, deep learning, and few-shot learning, which provided the possibility of recognizing the DAA of the grain timely. These results revealed that the ViT could improve the performance of deep learning in predicting the DAA, while few-shot learning could reduce the need for a number of datasets. This work provides a new approach to monitoring wheat grain filling dynamics, and it is beneficial for disaster prevention and improvement of wheat production.

Simultaneous prediction of wheat yield and grain protein content using multitask deep learning from time-series proximal sensing

Abstract

Wheat yield and grain protein content (GPC) are two main optimization targets for breeding and cultivation. Remote sensing provides nondestructive and early predictions of yield and GPC, respectively. However, whether it is possible to simultaneously predict yield and GPC in one model and the accuracy and influencing factors are still unclear. In this study, we made a systematic comparison of different deep learning models in terms of data fusion, time-series feature extraction, and multitask learning. The results showed that time-series data fusion significantly improved yield and GPC prediction accuracy with R2 values of 0.817 and 0.809. Multitask learning achieved simultaneous prediction of yield and GPC with comparable accuracy to the single-task model. We further proposed a two-to-two model that combines data fusion (two kinds of data sources for input) and multitask learning (two outputs) and compared different feature extraction layers, including RNN (recurrent neural network), LSTM (long short-term memory), CNN (convolutional neural network), and attention module. The two-to-two model with the attention module achieved the best prediction accuracy for yield (R2 = 0:833) and GPC (R2 = 0:846). The temporal distribution of feature importance was visualized based on the attention feature values. Although the temporal patterns of structural traits and spectral traits were inconsistent, the overall importance of both structural traits and spectral traits at the postanthesis stage was more important than that at the preanthesis stage. This study provides new insights into the simultaneous prediction of yield and GPC using deep learning from time-series proximal sensing, which may contribute to the accurate and efficient predictions of agricultural production.

Anti-gravity stem-seeking restoration algorithm for maize seed root image phenotype detection

Abstract

Root phenotype detection is key to cultivating seeds with excellent traits, and requires a complete root image. However, soil occlusion, uneven lighting, and other factors cause broken points and segments in the root image. To solve this problem, an anti-gravity stem-seeking (AGSS) root image restoration algorithm is proposed in this paper to repair root images and extract root phenotype information for different resistant maize seeds. First, the obtained root image was processed using uniform illumination, grayscale, binarization, and morphological filtering to separate it from the background. Subsequently, a root skeleton map was generated using a thinning algorithm for pixel-level image processing, and the Taproot junction G was obtained. Subsequently, root pixel coordinates were obtained by traversing the root skeleton map. All root-segment endpoint coordinates were obtained using the endpoint judgment rule and stored in the endpoint list. The lateral and primary root endpoints were separated based on the lateral root judgment rule, and stored in the side root and primary root endpoint lists, respectively. Subsequently, the primary root endpoints were processed and fitted in the order short to long using arbitrary-two-endpoint spacing until all breakpoints of the primary root were found. The coordinates of the top endpoints of each principal root were obtained. Finally, the top endpoint was connected to point G based on the Bezier curve-fitting method to achieve complete root repair. The proposed AGSS root image restoration algorithm was applied to detect the root systems of maize with different resistances and wheat to evaluate its performance against the standard dataset. The results indicated a detection accuracy of greater than 90% for root taproot length and diameter. It was also found that maize drought resistance was positively correlated with root length and diameter, but negatively with the lateral root number. In contrast, the waterlogging and salt resistance traits of maize were positively correlated with the number of lateral roots. In conclusion, the proposed AGSS root image restoration algorithm can quickly and effectively repair root images, is suitable for different resistance evaluations of maize seeds, and is conducive to the detection of root phenotypes. Compared with deep learning methods, this algorithm displays advantages of fast repair, low hardware platform requirements, and less requirement of training images. The algorithm is highly suitable for deploying in small embedded systems, with broad application prospects.