The rapid, non-destructive, and precise prediction of oil content in peanut seeds is of great significance for accelerating the breeding process and meeting the demands of the food industry. This study proposes a novel 3D weighted pooling and channel attention convolutional neural network (3D-WPCA-CNN) model to enhance the accuracy of oil content prediction. The model integrates two key components: a weighted average pooling layer and a channel attention mechanism, both designed to optimize feature extraction and information weighting in high-dimensional data. The weighted average pooling layer is developed to enhance the extraction of critical features while reducing interference from irrelevant or noisy information. Unlike conventional pooling methods, such as max pooling or average pooling, which treat all spatial features equally, this layer introduces learnable weight matrices that dynamically adjust based on feature importance. During the training process, these weights are continuously optimized, allowing the model to highlight the most informative spatial and spectral features while suppressing less relevant ones. The pooling operation is implemented using weighted 3D convolution, where each feature channel is assigned an independent weight matrix, ensuring that different spectral channels are processed distinctly. Additionally, the model adopts grouped convolution to maintain independence among channels and improve computational efficiency. The learnable weight parameters are initialized randomly and progressively fine-tuned through backpropagation, enabling the network to adaptively focus on key spectral and spatial information. The channel attention mechanism further enhances the model's feature selection capability by dynamically adjusting the importance of different channels in the hyperspectral image. Traditional convolutional neural networks (CNNs) treat all channels as equally important, which may not be optimal for hyperspectral data, where certain spectral bands carry more relevant information for predicting oil content. To address this, the proposed mechanism first applies global average pooling across the spatial dimensions to extract a channel-wise feature descriptor, summarizing the overall contribution of each channel. This descriptor is then processed through a lightweight fully connected network, which learns to assign importance weights to each channel. The network includes a hidden layer with a ReLU activation function, followed by an output layer with a sigmoid activation function, ensuring that the learned weights are normalized. The attention weights are then applied to reweight the input feature channels, emphasizing high-contribution channels while suppressing redundant or irrelevant ones. This process enables the model to refine its feature representation dynamically, improving its predictive accuracy. To validate the superiority of the proposed approach, the model was compared against a convolutional neural network (CNN) based on mean spectral features. The proposed 3D-CNN model demonstrated higher predictive accuracy (R² = 0.809 2, RMSE = 1.713 1%) compared to the model based on average spectra (R²= 0.698 3, RMSE = 2.183 7%), 3D Average pooling convolutional neural network without Channel Attention（R²=0.721 7, RMSE=2.069 3%）and the 3D Weighted pooling convolutional neural network without Channel Attention (R² = 0.738 4, RMSE = 2.005 9%). Furthermore, the RPD of the 3D CNN model was 2.338 9, indicating its reliable performance in predicting oil content. This study provides a novel high-dimensional modeling approach for hyperspectral image-based crop seed composition prediction, offering valuable practical implications for the precise assessment of quality traits in peanuts and other agricultural crops.

Traditional two-dimensional (2D) analyses of maize (Zea mays L.) plant shape plasticity and canopy transmission under varying planting densities have limitations in capturing spatial heterogeneity. This study used a three-dimensional (3D) phenotyping platform to investigate architectural plasticity across different maize varieties and planting densities. Seven novel 3D architectural parameters were developed, and 3D canopy models were constructed for light distribution simulation. At the vegetative stage 9 (V9) stage, medium planting density (67,500 plants ha^–1, MD) increased plant side width and convex hull volume by 7.2 and 11.4%, respectively, compared to low planting density (37,500 plants ha^–1, LD). High planting density (97,500 plants ha^–1, HD) increased the width and volume by 4.2 and 17.8%, respectively, compared with MD. Similar changes were maintained at the V13 stage. At the silking stage, the number of voxel volume plant (NVP) and projected area (PJA) decreased by 6.2 and 11.9%, respectively, under MD compared with LD, and by 4.9 and 3.6%, respectively, under HD compared with MD. Across all densities, PJA and NVP in both MC812 and JNK728 were consistently lower than in ZD958. A bottom light transmittance estimation model combining point cloud parameters with support vector regression achieved reliable predictions (R²=0.76, RMSE=2.89%). The 3D canopy model effectively simulated population light distribution (R²=0.83, RMSE=8.53%). NVP and PJA were identified as critical parameters affecting bottom canopy transmittance, suggesting their potential as 3D selection indices for maize density tolerance breeding. These findings provide insights into stage-specific architectural plasticity and light interception, supporting molecular design breeding of density-tolerant maize.

Identifying and segmenting the vitreous and starchy endosperm of maize kernels is essential for texture analysis. However, the complex internal structure of maize kernels presents several challenges. In CT (computed tomography) images, the pixel intensity differences between the vitreous and starchy endosperm regions in maize kernel CT images are not distinct, potentially leading to low segmentation accuracy or oversegmentation. Moreover, the blurred edges between the vitreous and starchy endosperm make segmentation difficult, often resulting in jagged segmentation outcomes. We propose a deep learning-based CT image analysis pipeline to examine the internal structure of maize seeds. First, CT images are acquired using a multislice CT scanner. To improve the efficiency of maize kernel CT imaging, a batch scanning method is used. Individual kernels are accurately segmented from batch-scanned CT images using the Canny algorithm. Second, we modify the conventional architecture for high-quality segmentation of the vitreous and starchy endosperm in maize kernels. The conventional U-Net is modified by integrating the CBAM (convolutional block attention module) mechanism in the encoder and the SE (squeeze-and-excitation attention) mechanism in the decoder, as well as by using the focal-Tversky loss function instead of the Dice loss, and the boundary smoothing term is weighted as an additional loss term, named CSFTU-Net. The experimental results show that the CSFTU-Net model significantly improves the ability of segmenting vitreous and starchy endosperm. Finally, a segmented mask-based method is proposed to extract phenotype parameters of maize kernel texture, including the volume of the kernel (V), volume of the vitreous endosperm (VV), volume of starchy endosperm (SV), and ratios over their respective total kernel volumes (VV/V and SV/V). The proposed pipeline facilitates the nondestructive quantification of the internal structure of maize kernels, offering valuable insights for maize breeding and processing.

Marked variations in the 3-dimensional (3D) shape of corn leaves can be discerned as a function of various influences, including genetics, environmental factors, and the management of cultivation processes. However, the causes of these variations remain unclear, primarily due to the absence of quantitative methods to describe the 3D spatial morphology of leaves. To address this issue, this study acquired 3D digitized data of ear-position leaves from 478 corn inbred lines during the grain-filling stage. We propose quantitative calculation methods for 13 3D leaf shape features, such as the leaf length, 3D leaf area, leaf inclination angle, blade-included angle, blade self-twisting, blade planarity, and margin amplitude. Correlation analysis, cluster analysis, and heritability analysis were conducted among the 13 leaf traits. Leaf morphology differences among subpopulations of the inbred lines were also analyzed. The results revealed that the 3D leaf traits are capable of revealing the morphological differences among different leaf surfaces, and the genetic analysis revealed that 84.62% of the 3D phenotypic traits of ear-position leaves had a heritability greater than 0.3. However, the majority of 3D leaf shape traits were strongly affected by environmental conditions. Overall, this study quantitatively investigated 3D leaf shape in corn, providing a reliable basis for further research on the genetic regulation of corn leaf morphology and advancing the understanding of the complex interplay among crop genetics, phenotypes, and the environment.

The radiation use efficiency (RUE) is one of the most important functional traits determining crop productivity. The coordination of the vertical distribution of light and leaf nitrogen has been proven to be effective in boosting the RUE from both experimental and computational evidence. However, previous simulation studies have primarily assumed that the leaf area is uniformly distributed along the canopy depth, rarely considering the optimization of the leaf area distribution, especially for C4 crops. The present study hypothesizes that the RUE may be maximized by matching the leaf area and leaf nitrogen vertical distributions in the canopy. To test this hypothesis, various virtual maize canopies were generated by combining the leaf inclination angle, vertical leaf area distribution, and vertical leaf nitrogen distribution and were further evaluated by an improved multilayer canopy photosynthesis model. We found that a greater fraction of leaf nitrogen is preferentially allocated to canopy layers with greater leaf areas to maximize the RUE. The coordination of light and nitrogen emerged as a property from the simulations to maximize the RUE in most scenarios, particularly in dense canopies. This study not only facilitates explicit and precise profiling of ideotypes for maximizing the RUE but also represents a primary step toward high-throughput phenotyping and screening of the RUE for massive numbers of inbred lines and cultivars.

The tassel state in maize hybridization fields not only reflects the growth stage of the maize but also reflects the performance of the detasseling operation. Existing tassel detection models are primarily used to identify mature tassels with obvious features, making it difficult to accurately identify small tassels or detasseled plants. This study presents a novel approach that utilizes unmanned aerial vehicles (UAVs) and deep learning techniques to accurately identify and assess tassel states, before and after manually detasseling in maize hybridization fields. The proposed method suggests that a specific tassel annotation and data augmentation strategy is valuable for substantial enhancing the quality of the tassel training data. This study also evaluates mainstream object detection models and proposes a series of highly accurate tassel detection models based on tassel categories with strong data adaptability. In addition, a strategy for blocking large UAV images, as well as improving tassel detection accuracy, is proposed to balance UAV image acquisition and computational cost. The experimental results demonstrate that the proposed method can accurately identify and classify tassels at various stages of detasseling. The tassel detection model optimized with the enhanced data achieves an average precision of 94.5% across all categories. An optimal model combination that uses blocking strategies for different development stages can improve the tassel detection accuracy to 98%. This could be useful in addressing the issue of missed tassel detections in maize hybridization fields. The data annotation strategy and image blocking strategy may also have broad applications in object detection and recognition in other agricultural scenarios.

The 3-dimensional (3D) modeling of crop canopies is fundamental for studying functional-structural plant models. Existing studies often fail to capture the structural characteristics of crop canopies, such as organ overlapping and resource competition. To address this issue, we propose a 3D maize modeling method based on computational intelligence. An initial 3D maize canopy is created using the t-distribution method to reflect characteristics of the plant architecture. The subsequent model considers the 3D phytomers of maize as intelligent agents. The aim is to maximize the ratio of sunlit leaf area, and by iteratively modifying the azimuth angle of the 3D phytomers, a 3D maize canopy model that maximizes light resource interception can be constructed. Additionally, the method incorporates a reflective approach to optimize the canopy and utilizes a mesh deformation technique for detecting and responding to leaf collisions within the canopy. Six canopy models of 2 varieties plus 3 planting densities was constructed for validation. The average R² of the difference in azimuth angle between adjacent leaves is 0.71, with a canopy coverage error range of 7% to 17%. Another 3D maize canopy model constructed using 12 distinct density gradients demonstrates the proportion of leaves perpendicular to the row direction increases along with the density. The proportion of these leaves steadily increased after 9 × 10⁴ plants ha⁻¹. This study presents a 3D modeling method for the maize canopy. It is a beneficial exploration of swarm intelligence on crops and generates a new way for exploring efficient resources utilization of crop canopies.

The field phenotyping platforms that can obtain high-throughput and time-series phenotypes of plant populations at the 3-dimensional level are crucial for plant breeding and management. However, it is difficult to align the point cloud data and extract accurate phenotypic traits of plant populations. In this study, high-throughput, time-series raw data of field maize populations were collected using a field rail-based phenotyping platform with light detection and ranging (LiDAR) and an RGB (red, green, and blue) camera. The orthorectified images and LiDAR point clouds were aligned via the direct linear transformation algorithm. On this basis, time-series point clouds were further registered by the time-series image guidance. The cloth simulation filter algorithm was then used to remove the ground points. Individual plants and plant organs were segmented from maize population by fast displacement and region growth algorithms. The plant heights of 13 maize cultivars obtained using the multi-source fusion data were highly correlated with the manual measurements (R² = 0.98), and the accuracy was higher than only using one source point cloud data (R² = 0.93). It demonstrates that multi-source data fusion can effectively improve the accuracy of time series phenotype extraction, and rail-based field phenotyping platforms can be a practical tool for plant growth dynamic observation of phenotypes in individual plant and organ scales.