Computers and Electronics in Agriculture最新文献

Trajectory data acquired from GNSS (Global Navigation Satellite System) terminals on agricultural machinery are crucial for identifying agricultural machinery operation modes, evaluating agricultural machinery operational efficiency and exploring agricultural machinery trans-regional harvesting operation characteristics. However, GNSS terminals often experience signal delays due to factors such as weather conditions and environmental obstructions. These delays result in irregular time intervals between trajectory points, leading to local sparsity within the trajectory data, which subsequently reduces the accuracy of applications and analyses based on agricultural machinery trajectories. To address this issue, we propose a novel approach that leverages Bidirectional Long Short-Term Memory (BiLSTM) and Gated Recurrent Unit (GRU) networks, along with an attention mechanism, to mitigate the problem of local trajectory sparsity, and experiments were conducted using agricultural machinary trajectory data collected during the 2023 wheat harvest period. The results demonstrate the efficiency of our approach by successfully resolving the local sparsity of agricultural machinery trajectories. Moreover, each newly added trajectory point contains all original attributes (e.g., speed and direction). When integrated into state-of-the-art algorithms (e.g., DT, DBSCAN + rules, GCN) for identifying machinery operation modes, our method improves accuracies by 21.83 %, 26.86 %, and 1.17 %, respectively. Our approach effectively addresses the issue of local trajectory sparsity, thus providing assistance for applications and studies based on massive agricultural machinery trajectories.

Accurate crop yield estimation is essential for ensuring national food security and sustainable development. The crop growth model is one of the primary methods for estimating production and has been effectively applied in simulating crop growth and development, environmental factor effects and yield formation processes. However, many related studies have focused on Gramineae crops, such as wheat, rice and maize, while there has been little research on rape and soybean yield estimation. Non-foliar green organs such as rape siliques and soybean pods make vital contributions to plant photosynthesis and influence crop output. The parameter calibration method based on the leaf area index (LAI) cannot satisfy the existing demand for high-precision yield estimation for crops with active non-foliar green organs. Therefore, a new method for crop growth model calibration was proposed, which considers non-foliar green organs and plant photosynthetic succession processes and combines them with those of leaves to construct a photosynthetic area index (PAI). With wheat, rape and soybean as the research objects, their yields were simulated via the proposed crop growth model calibration method. The results showed that using the proposed calibration method to calibrate World Food Study (WOFOST) model parameters on the basis of the PAI improved the yield estimation accuracy over that of the crop model calibration method based on the LAI. The determination coefficient (R²) of the total dry weight of storage organs (TWSO) simulation value increased from 0.73 (using the LAI) to more than 0.90 (using the PAI). The R² values of TWSO at the rape calibration and verification points were 0.910 and 0.922, respectively, and the R² values of TWSO at the soybean calibration and verification points were 0.741 and 0.926, respectively. The above results verified the feasibility and effectiveness of the proposed calibration method. Consequently, the application of the crop model calibration method proposed in this paper is important for accurately estimating the crop yield of plants with active non-foliar green organs, promoting the expansion of general crop models for different types of crops and achieving high-precision crop yield estimations.

In recent years, there have been relevant advances in precision beekeeping. These advances are mainly focused on proposing sensor systems that collect crucial information for bee welfare, creating integrated architectures that allow beekeepers to monitor the current state of their hive through real-time data. However, there is a lack of predictive models that would allow beekeepers to anticipate specific events that endanger bee welfare and lead to a decline in productivity. Specifically, predictive approaches accounting for the high correlation among internal variables of beehives have not been developed to date. To address this research gap, multivariate predictive models, including auto-regressive state-space and time series models, have been implemented and applied to four different hives from the we4bee project. These models aim to predict the internal variables of beehives (four different temperatures, humidity, and weight) by utilizing the meteorological conditions to which the hives are exposed. A cross-validation adapted to time series data was employed for model generalization assessment. Prediction models based on vector time series exhibited superior performance in forecasting internal hive variables compared to multivariate auto-regressive state-space models. Overall, the approach based on the vector error correction model yielded the best balance between fit, prediction, and computational cost. The VEC-based approach produces predictions with maximum mean absolute errors of 177 $\left(312\right)$g in weight, 3.366 $(3.802)\text{\%}$ in humidity, and 1.122 $(1.685)°C$ in temperature at 1 (3)-days ahead when dealing with beehives exhibiting a high degree of correlation in their internal variables. Moreover, the VEC-based approach requires less than a second to perform the time series fitting process, which makes it particularly interesting for application in big data environments. The integration of such models into a decision support system would meet the need of beekeepers to anticipate potential threats to the welfare of their bee colonies, streamlining their monitoring processes while eliminating the need for continuous inspections.

The precise monitoring of fish keypoints and behavioral patterns is crucial in fish farming, influencing decisions on feeding schedules and assessing fish health. Traditional approaches to multi-object pose estimation often lean towards either Bottom-up or Top-down methods. Our innovative solution introduces a streamlined single-stage multi-object pose estimation framework, utilizing a unique approach to instance assignment based on location. By incorporating position encoding as an index for candidate pose heatmap groups, we achieve end-to-end multi-object pose estimation with reduced redundancy through non-maximum suppression. Our framework, named POLO, has been validated on a meticulously annotated fish keypoint dataset, demonstrating outstanding performance with a remarkable 65.34% OKS AP at 71.4 FPS on Tesla v100. With its real-time capabilities, POLO is highly adaptable, making it suitable for deployment on various edge computing devices and addressing real-world challenges effectively. We believe our framework can serve as a solid baseline for diverse pose estimation tasks across different domains.

In this study, we propose a novel visual-based autonomous trajectory-tracking control method for steering a wheeled robot following the lines of crop row in paddy field. A rice crop rows detection method, based on the region growth sequential clustering − random sample consensus (RANSAC) algorithm, is developed to generate trajectory. Concurrently, a dynamics predictive controller is employed to compute the command for the desired steering angle. The controller leverages a model that incorporates slip dynamics and operates on a low power consumption industrial computer. Experimental results show that the developed algorithm can successfully obtain the correct trajectory in more than 96.25 % of the cases, with the angle error consistently below 3°. Furthermore, the single-image processing time is notably swift at 13.98 ms, underscoring the commendable adaptability and real-time performance of the proposed methodology. During movement in the paddy field, the robot exhibits maximum lateral deviations of 4.55 cm, 5.65 cm, and 6.41 cm at speeds of 0.3 m/s, 0.6 m/s, and 0.9 m/s, respectively, accompanied by corresponding heading angle errors of 4.59°, 5.63°, and 7.39°. Notably, while adeptly tracking rice crop rows at all three speeds, the robot consistently maintains a maximum lateral error below one-fourth of the inter-row spacing of rice planting. This study assumes significance in enhancing the stability of ground-traversing agricultural robots, serving as a valuable reference for advancing the research and development of intelligent and efficient agricultural robotic systems.

The broken rate and impurity rate of wheat are important indicators for assessing the quality of combine harvester operations. In view of the overlapping, occlusion and dense adhesion between the scattered grains during the operation of the combine harvester, it is difficult to obtain the grain crushing characteristics and impurity mass, which leads to low detection accuracy. In this paper, a method for detecting wheat broken rate and impurity rate based on DeepLab-EDA semantic segmentation model was proposed, and a detection system was built. In the detection system, an image acquisition device was designed and developed based on the principle of electromagnetic vibration, and the deep learning model was deployed in the embedded processor. Through the human–computer interaction interface design, the online processing and analysis of wheat image data and the display of the detection results of broken rate and impurity rate were realized. Comparative experiments with traditional semantic segmentation models showed that the MIoU, MP and MR of the DeepLab-EDA model were 89.41 %, 95.97 % and 94.83 %, respectively, representing improvements of 9.94%, 7.41%, and 7.52% over the baseline model, and indicating a significant enhancement in the accurate identification and segmentation of broken grain and impurities. Based on this, indoor group matching experiments were conducted with three groups of broken rate and impurity rate levels set at 0.5%, 1.5%, and 2.5%, showing the average errors of 7.54% and 6.30% for broken rate and impurity rate detection systems, respectively. Furthermore, the detection device was installed under the grain outlet of the GM80 combine harvester for field experiments, which showed average errors of 13.32% and 9.77% for wheat broken rate and impurity rate, respectively. The effectiveness and accuracy of the wheat broken rate and impurity rate detection system meet the requirements, which can provide a data basis for intelligent control of combine harvester operation parameters by the operator.

The automatic acquisition of crop information can promote the rapid development of precision agriculture. Tomatoes, as a representative greenhouse crop, can demonstrate their growth status and overall health could be exhibited by measuring the diameters of their main stems. Automated measurement of the diameters of the trunk stems can not only reduce labor costs, but also enhance efficiency, thereby facilitating the improvement of tomato cultivation and planting management. Therefore, this study proposes a novel method based on an instance segmentation algorithm combined with RGB-D data to measure the diameter of a tomato trunk. First, we utilize the improved YOLOv8-seg to acquire the masks and bounding boxes of the buds and stems. Namely, we replace the SPPF module with Soft-SPPF to enhance the model’s capability to extract multi-scale features. Additionally, we improve the neck layer using cross-stage and weighted feature fusion to enhance the feature fusion effect. Then, we design a method to calculate the diameter of the tomato trunk by utilizing the instance segmentation results and depth information. Specifically, we first obtain the measurement point and the ROI (Region of Interest) to be measured by identifying the intersection between the bounding box of the bud and the straight line fitted by the stem mask in an image. We then filter out irrelevant depth information using the mask within the ROI, optimize the coordinate values of the measurement point, and calculate the main stem diameter. The results demonstrate that the measurements obtained in this study have a RMSE (Root Mean Square Error) of 1.5 mm and a MAPE (Mean Absolute Percentage Error) of 12.37 % compared with the manual measurements. Compared with the method based on object detection, the direct acquisition of contour information for the target instances by the instance segmentation algorithm reduces the algorithmic complexity of the subsequent processing. The proposed method can offer precise and reliable information on the diameter of the main stem of a tomato in real-world scenarios. It can be extended to other types of crops in greenhouses.

The high-density culture of the recirculating aquaculture system (RAS) makes bacterial and parasitic diseases in fish likely to spread within these systems. Therefore, preventing and controlling the occurrence of fish diseases in RAS is crucial for the future development of aquaculture. However, the current circulating water system lacks early warning measures to prevent and control diseases, resulting in poor accuracy in early warning and detection of diseases. To tackle this challenge, this study introduces an early warning system for largemouth bass (Micropterus salmoides) disease, which is based on You Only Look Once vision 8 (YOLOv8), ByteTrack, Long-Short-Term-Memory (LSTM), and Fuzzy Inference System (FIS). The system utilizes the water quality, surface characteristics, and behavioral traits of diseased fish to predict and prevent disease outbreaks. The system achieved an accuracy of 79.33% for identifying infected body surface features, 80.65% for identifying diseased water quality, and 81.08% for predicting diseased behavior. The experimental results indicate that the early warning system is highly reliable and effective, achieving integrated disease identification accuracy as high as 94.08%. This study enhances the accuracy of early disease warning in fish disease conditions, achieving early warning of nocardiosis in largemouth bass. The study provides crucial technical support for the sustainable and high-quality development of the aquaculture industry.

To enhance urban forestry efficacy in Hong Kong, implementing a paradigm shift towards an automated urban tree inventory that utilizes advanced sensing technologies and artificial intelligence is essential for streamlined data collection and analysis. This study advances this objective by creating a comprehensive framework for estimating diameter at breast height (DBH) and extracting tree images. This framework encompasses five key stages: (1) data acquisition utilizing StructXray, a mobile mapping system equipped with a 360° camera and a multi-beam flash LiDAR sensor; (2) vegetation point clouds extraction using deep learning techniques; (3) individual tree segmentation through machine learning algorithms; (4) DBH estimation; and (5) tree image extraction. Six datasets were collected, yielding tree detection precision, recall and F1 score of 0.88, 0.95 and 0.91 respectively. The presence of moving objects within the 3D point cloud map, exhibiting diverse geometric structures, hinders precise vegetation point cloud segmentation by the pointwise neural network. To tackle this challenge, SalsaNext was employed to rectify the predictions of a pointwise neural network, specifically RandLA-Net in this study, eliminating 91 % of misclassified moving object point clouds and completely removing them from 47 % of affected individual tree point clouds. Additionally, a chord length-based method was proposed to enhance DBH estimation accuracy by dividing the point cloud slice into sectors and summing the chord lengths to estimate the tree trunk perimeter. Compared to the ellipse least squares fitting method, this approach reduced the root-mean-square error of the estimated DBH by 1.31 cm.

Accurate detection and differentiation of grapevine canopies from other vegetation, along with individual grapevine row identification, pose significant challenges in precision viticulture (PV), especially within irregularly structured vineyards shaped by natural terrain slopes. This study employs aerial imagery captured by unmanned aerial vehicles (UAVs) and introduces an image processing methodology that relies on the orthorectified raster data obtained through UAVs. The proposed method adopts a data-driven approach that combines visible indices and elevation data to achieve precise grapevine row detection. Thoroughly tested across various vineyard configurations, including irregular and terraced landscapes, the findings underscore the method’s effectiveness in identifying grapevine rows of diverse shapes and configurations. This capability is crucial for accurate vineyard monitoring and management. Furthermore, the method enables clear differentiation between inter-row spaces and grapevine vegetation, representing a fundamental advancement for comprehensive vineyard analysis and PV planning. This study contributes to the field of PV by providing a reliable tool for grapevine row detection and vineyard feature classification. The proposed methodology is applicable to vineyards with varying layouts, offering a versatile solution for enhancing precision viticulture practices.