The climatic parameters within greenhouse facilities, such as temperature, humidity, and light, exert significant influence on the growth and yield of crops, particularly seedlings. Therefore, it is crucial to establish an accurate predictive model to monitor and adjust the greenhouse microclimate for optimizing the greenhouse environment to the fullest extent. To precisely forecast the greenhouse microclimate and assess the suitability of nursery environments, this study focuses on greenhouse environmental factors. This study leveraged open-source APIs to acquire meteorological data, integrated a model based on Convolutional Neural Networks (CNN) and Long Short-Term Memory Networks (LSTM), and utilized the sparrow search algorithm to optimize model parameters, consequently developing a time series greenhouse microclimate prediction model. Furthermore, Squeeze-and-Excitation (SE) Networks were employed to enhance the model’s attention mechanism, enabling more accurate predictions of environmental factors within the greenhouse. The predictive results indicated that the optimized model achieved high precision in forecasting the greenhouse microclimate, with average errors of 0.540 °C, 0.936%, and 1.586 W/m² for temperature, humidity, and solar radiation, respectively. The coefficients of determination (${\mathrm{R}}^2$) reached 0.940, 0.951, and 0.936 for temperature, humidity, and solar radiation, respectively. In comparison to individual CNN or LSTM models, as well as the back-propagation (BP) neural network, the proposed model demonstrates a significant improvement in predictive accuracy. Moreover, this research was applied to the greenhouse nursery environment, demonstrating that the proposed model significantly enhanced the efficiency of greenhouse seedling cultivation and the quality of seedlings. Our study provided an effective approach for optimizing greenhouse environmental control and nursery environment suitability, contributing significantly to achieving sustainable and efficient agricultural production.

Peanut leaf spot is a worldwide disease whose prevalence poses a major threat to peanut yield and quality, and accurate prediction models are urgently needed for timely disease management. In this study, we proposed a novel peanut leaf spot prediction method based on an improved long short-term memory (LSTM) model and multi-year meteorological data combined with disease survey records. Our method employed a combination of convolutional neural networks (CNNs) and LSTMs to capture spatial–temporal patterns from the data and improve the model’s ability to recognize dynamic features of the disease. In addition, we introduced a Squeeze-and-Excitation (SE) Network attention mechanism module to enhance model performance by focusing on key features. Through several hyper-parameter optimization adjustments, we identified a peanut leaf spot disease condition index prediction model with a learning rate of 0.001, a number of cycles (Epoch) of 800, and an optimizer of Adma. The results showed that the integrated model demonstrated excellent prediction ability, obtaining an RMSE of 0.063 and an $R^2$ of 0.951, which reduced the RMSE by 0.253 and 0.204, and raised the $R^2$ by 0.155 and 0.122, respectively, compared to the single CNN and LSTM. Predicting the occurrence and severity of peanut leaf spot disease based on the meteorological conditions and neural networks is feasible and valuable to help growers make accurate management decisions and reduce disease impacts through optimal fungicide application timing.

Grapes are a globally popular fruit, with grape cultivation worldwide being second only to citrus. This article focuses on the low efficiency and accuracy of traditional manual grading of red grape external appearance and proposes a small-sample red grape external appearance grading model based on transfer learning with convolutional neural networks (CNNs). In the naturally growing vineyards, 195,120,135 samples of Grade I, Grade II, and Grade III red grapes were collected using a Canon EOS 550D camera, and a data set of 1800 samples was obtained using data enhancement technology. Then, the CNN transfer learning method was used to transfer the pre-trained AlexNet, VGG16, GoogleNet, InceptionV3, and ResNet50 network models on the ImageNet image dataset to the red grape image grading task. By comparing the classification performance of the CNN models of these five different network depths with fine-tuning, ResNet50 with a learning rate of 0.001 and a loop number of 10 was determined to be the best feature extractor for red grape images. Moreover, given the small number of red grape image samples in this study, different convolutional layer features output by the ResNet50 feature extractor were analyzed layer by layer to determine the effect of deep features extracted by each convolutional layer on Support Vector Machine (SVM) classification performance. This analysis helped to obtain a ResNet50 + SVM red grape external appearance grading model based on the optimal ResNet50 feature extraction strategy. Experimental data showed that the classification model constructed using the feature parameters extracted from the 10th node of the ResNet50 network achieved an accuracy rate of 95.08% for red grape grading. These research results provide a reference for the online grading of red grape clusters based on external appearance quality and have certain guiding significance for the quality and efficiency of grape industry circulation and production.

Tomato blight significantly threatened tomato yield and quality, making precise disease detection essential for modern agricultural practices. Traditional segmentation models often struggle with over-segmentation and missed segmentation, particularly in complex backgrounds and with diverse lesion morphologies. To address these challenges, we proposed Unet with Vision Mamba and ConvNeXt (VMC-Unet), an asymmetric segmentation model for quantitative analysis of tomato blight. Built on the Unet framework, VMC-Unet integrated a parallel feature-aware backbone combining ConvNeXt, Vision Mamba, and Atrous Spatial Pyramid Pooling (ASPP) modules to enhance spatial feature focusing and multi-scale information processing. During decoding, Vision Mamba was hierarchically embedded to accurately recover complex lesion morphologies through refined feature processing and efficient up-sampling. A joint loss function was designed to optimize the model’s performance. Extensive experiments on both tomato epidemic and public datasets demonstrated VMC-Unet superior performance, achieving 97.82% pixel accuracy, 87.94% F1 score, and 86.75% mIoU. These results surpassed those of classical segmentation models, underscoring the effectiveness of VMC-Unet in mitigating over-segmentation and under-segmentation while maintaining high segmentation accuracy in complex backgrounds. The consistent performance of the model across various datasets further validated its robustness and generalization potential, highlighting its applicability in broader agricultural settings.

Quantitative phenotyping of pepper seedlings is important for greenhouse plug tray seedling cultivation, but it remains constrained by inefficient manual monitoring, complex greenhouse backgrounds, and growth-stage-dependent discrepancies between two-dimensional image traits and actual leaf biomass. In this study, a cascaded vision framework with stage-specific morphological correction was developed for nondestructive seedling phenotyping. The framework integrated Visual Dynamic Momentum YOLO (VDM-YOLO) for individual seedling localization and growth-stage recognition, Variance Guided Strip Ghost Gated UNet (VSG-UNet) for lightweight, high-resolution leaf segmentation, and a stage-aware correction model for leaf dry biomass estimation. In performance evaluation, VDM-YOLO achieved a mean average precision at an intersection over union threshold of 0.5 (mAP_0.5) of 89.27%, improving mAP_0.5 by 1.82 percentage points over YOLOv12. VSG-UNet achieved a mean intersection over union (mIoU) of 83.9% and a Dice coefficient of 81.8%, while reducing floating point operations (FLOPs) and parameters by 44.2% and 61.2%, respectively, compared with U-Net. After stage-aware calibration, the coefficient of determination (R²) between segmented area and leaf dry weight increased from 0.764 to 0.813, and the root mean square error (RMSE) decreased from 0.0210 g to 0.0190 g. These results demonstrated that the proposed framework provided a proof of concept approach based on RGB images for the nondestructive assessment of leaf area and leaf dry biomass in pepper seedlings under restricted experimental conditions.