With the advancement of computer vision technology, the demand for fruit recognition in agricultural automation is increasing. To improve the accuracy and efficiency of recognizing young red pears, this study proposes an improved model based on the lightweight YOLOv9s, termed YOLOv9s-Pear. By constructing a feature-rich and diverse image dataset of young red pears and introducing spatial-channel decoupled downsampling (SCDown), C2FUIBELAN, and the YOLOv10 detection head (v10detect) modules, the YOLOv9s model was enhanced to achieve efficient recognition of small targets in resource-constrained agricultural environments. Images of young red pears were captured at different times and locations and underwent preprocessing to establish a high-quality dataset. For model improvements, this study integrated the general inverted bottleneck blocks from C2f and MobileNetV4 with the RepNCSPELAN4 module from the YOLOv9s model to form the new C2FUIBELAN module, enhancing the model’s accuracy and training speed for small-scale object detection. Additionally, the SCDown and v10detect modules replaced the original AConv and detection head structures of the YOLOv9s model, further improving performance. The experimental results demonstrated that the YOLOv9s-Pear model achieved high detection accuracy in recognizing young red pears, while reducing computational costs and parameters. The detection accuracy, recall, mean precision, and extended mean precision were 0.971, 0.970, 0.991, and 0.848, respectively. These results confirm the efficiency of the SCDown, C2FUIBELAN, and v10detect modules in young red pear recognition tasks. The findings of this study not only provide a fast and accurate technique for recognizing young red pears but also offer a reference for detecting young fruits of other fruit trees, significantly contributing to the advancement of agricultural automation technology.

The identification and enumeration of peach seedling fruits are pivotal in the realm of precision agriculture, greatly influencing both yield estimation and agronomic practices. This study introduces an innovative, lightweight YOLOv8 model for the automatic detection and quantification of peach seedling fruits, designated as YOLO-Peach, to bolster the scientific rigor and operational efficiency of orchard management. Traditional identification methods, which are labor-intensive and error-prone, have been superseded by this advancement. A comprehensive dataset was meticulously curated, capturing the rich characteristics and diversity of peach seedling fruits through high-resolution imagery at various times and locations, followed by meticulous preprocessing to ensure data quality. The YOLOv8s model underwent a series of lightweight optimizations, including the integration of MobileNetV3 as its backbone, the p2BiFPN architecture, spatial and channel reconstruction convolution, and coordinate attention mechanism, all of which have significantly bolstered the model’s capability to detect small targets with precision. The YOLO-Peach model excels in detection accuracy, evidenced by a precision and recall of 0.979, along with an mAP50 of 0.993 and an mAP50-95 of 0.867, indicating its superior capability for peach sapling identification with efficient computational performance. The findings underscore the model’s efficacy and practicality in the context of peach seedling fruit recognition. Ablation studies have shed light on the indispensable role of each component, with MobileNetV3 streamlining the model’s complexity and computational load, while the p2BiFPN architecture, ScConv convolutions, and coordinate attention mechanism have collectively enhanced the model’s feature extraction and detection precision for minute targets. The implications of this research are profound, offering a novel approach to peach seedling fruit recognition and serving as a blueprint for the identification of young fruits in other fruit species. This work holds significant theoretical and practical value, propelling forward the broader field of agricultural automation.

Maize is one of the important grain crops grown globally, and growth will directly affect its yield and quality, so it is important to monitor maize growth efficiently and non-destructively. To facilitate the use of unmanned aerial vehicles (UAVs) for maize growth monitoring, comprehensive growth indicators for maize monitoring based on multispectral remote sensing imagery were established. First of all, multispectral image data of summer maize canopy were collected at the jointing stage, and meanwhile, leaf area index (LAI), relative chlorophyll content (SPAD), and plant height (VH) were measured. Then, the comprehensive growth monitoring indicators CGMI_CV and CGMI_CR for summer maize were constructed by the coefficient of variation method and the CRITIC weighting method. After that, the CGMI_CV and CGMI_CR prediction models were established by the partial least-squares (PLSR) and sparrow search optimization kernel extremum learning machine (SSA-KELM) using eight typical vegetation indices selected. Finally, a comparative analysis was performed using ground-truthing data, and the results show: (1) For CGMI_CV, the R² and RMSE of the model built by SSA-KELM are 0.865 and 0.040, respectively. Compared to the model built by PLSR, R² increased by 4.5%, while RMSE decreased by 0.3%. For CGMI_CR, the R² and RMSE of the model built by SSA-KELM are 0.885 and 0.056, respectively. Compared to the other model, R² increased by 4.6%, and RMSE decreased by 2.8%. (2) Compared to the models by single indicator, among the models constructed based on PLSR, the CGMI_CR model had the highest R². In the models constructed based on SSA-KELM, the R² of models by the CGMI_CR and CGMI_CV were larger than that of the models by SPAD (R² = 0.837), while smaller than that of the models by LAI (R² = 0.906) and models by VH (R² = 0.902). In summary, the comprehensive growth monitoring indicators prediction model established in this paper is effective and can provide technical support for maize growth monitoring.

UAV (unmanned aerial vehicle) remote sensing provides the feasibility of high-throughput phenotype nondestructive acquisition at the field scale. However, accurate remote sensing of crop physicochemical parameters from UAV optical measurements still needs to be further studied. For this purpose, we put forward a crop phenotype inversion framework based on the optimal estimation (OE) theory in this paper, originating from UAV low-altitude hyperspectral/multispectral data. The newly developed unified linearized vector radiative transfer model (UNL-VRTM), combined with the classical PROSAIL model, is used as the forward model, and the forward model was verified by the wheat canopy reflectance data, collected using the FieldSpec Handheld in Qi County, Henan Province. To test the self-consistency of the OE-based framework, we conducted forward simulations for the UAV multispectral sensors (DJI P4 Multispectral) with different observation geometries and aerosol loadings, and a total of 801 sets of validation data were obtained. In addition, parameter sensitivity analysis and information content analysis were performed to determine the contribution of crop parameters to the UAV measurements. Results showed that: (1) the forward model has a strong coupling between vegetation canopy and atmosphere environment, and the modeling process is reasonable. (2) The OE-based inversion framework can make full use of the available radiometric spectral information and had good convergence and self-consistency. (3) The UAV multispectral observations can support the synchronous retrieval of LAI (leaf area index) and Cab (chlorophyll a and b content) based on the proposed algorithm. The proposed inversion framework is expected to be a new way for phenotypic parameter extraction of crops in field environments and had some potential and feasibility for UAV remote sensing.

Chlorophyll a and b content (Cab) and leaf area index (LAI) are two key parameters of crops, and their quantitative inversions are important for growth monitoring and the field management of wheat. However, due to the close correlation between the spectral signals of these two parameters and the effects of soil and atmospheric conditions, as well as modeling errors, synchronous retrieval of LAI and Cab from remote sensing data is still a challenging task. In a previous study, we introduced the optimal estimation theory and established the inversion framework by coupling the PROSAIL (PROSPECT + SAIL) model with the unified linearized vector radiative transfer model (UNL-VRTM). The framework fully utilizes the simulated radiance spectra for synchronous retrieval of Cab and LAI at the UAV observation scale and has good convergence and self-consistency. In this study, based on this inversion framework, synchronized retrieval of Cab and LAI was carried out by real wheat UAV observation data and validated with the ground-measured data. By comparing with the empirical statistical model constructed by the PROSAIL model and coupled model, least squares support vector machine (LSSVM), and random forest (RF), the proposed method has the highest accuracy of Cab and LAI estimated from UAV multispectral data (for Cab, R² = 0.835, RMSE = 14.357; for LAI, R² = 0.892, RMSE = 0.564). Our proposed method enables the fast and efficient estimation of Cab and LAI in multispectral data without prior measurements and training.

Wheat spike detection holds significant importance for agricultural production as it enhances the efficiency of crop management and the precision of operations. This study aims to improve the accuracy and efficiency of wheat spike detection, enabling efficient crop monitoring under resource-constrained conditions. To this end, a wheat spike dataset encompassing multiple growth stages was constructed, leveraging the advantages of MobileNet and ShuffleNet to design a novel network module, SeCUIB. Building on this foundation, a new wheat spike detection network, LGWheatNet, was proposed by integrating a lightweight downsampling module (DWDown), spatial pyramid pooling (SPPF), and a lightweight detection head (LightDetect). The experimental results demonstrate that LGWheatNet excels in key performance metrics, including Precision, Recall, and Mean Average Precision (mAP50 and mAP50-95). Specifically, the model achieved a Precision of 0.956, a Recall of 0.921, an mAP50 of 0.967, and an mAP50-95 of 0.747, surpassing several YOLO models as well as EfficientDet and RetinaNet. Furthermore, LGWheatNet demonstrated superior resource efficiency with a parameter count of only 1,698,529 and GFLOPs of 5.0, significantly lower than those of competing models. Additionally, when combined with the Slicing Aided Hyper Inference strategy, LGWheatNet further improved the detection accuracy of wheat spikes, especially for small-scale targets and edge regions, when processing large-scale high-resolution images. This strategy significantly enhanced both inference efficiency and accuracy, making it particularly suitable for image analysis from drone-captured data. In wheat spike counting experiments, LGWheatNet also delivered exceptional performance, particularly in predictions during the filling and maturity stages, outperforming other models by a substantial margin. This study not only provides an efficient and reliable solution for wheat spike detection but also introduces innovative methods for lightweight object detection tasks in resource-constrained environments.