Search

This study proposes an improved link prediction model for predicting the “suitable for people” relationship within the knowledge graph of tea. The relationships between various types of tea and suitable target groups have yet to be fully explored, and the existing InteractE model still does not adequately capture a portion of the complex information around the interactions between entities and relationships. In this study, we integrate SENet into the feature layer of the InteractE model to enhance the capturing of helpful information in the feature channels. Additionally, the GCN layer is employed as the encoder, and the SENet-integrated InteractE model is used as the decoder to further capture the neighbour node information in the knowledge graph. Furthermore, our proposed improved model demonstrates significant improvements compared to several standard models, including the original model from public datasets (WN18RR, Kinship). Finally, we construct a tea dataset comprising 6698 records, including 330 types of tea and 29 relationship types. We predict the “suitable for people” relationship in the tea dataset through transfer learning. When comparing our model with the original model, we observed an improvement of 1.4% in H@10 for the WN18RR dataset, a 7.6% improvement in H@1 for the Kinship dataset, and a 5.2% improvement in MRR. Regarding the tea dataset, we achieved a 4.1% increase in H@3 and a 2.5% increase in H@10. This study will help to fully exploit the value potential of tea varieties and provide a reference for studies assessing healthy tea drinking.

Accurate temperature prediction and modeling are critical for effective management of agricultural greenhouses. By optimizing control and minimizing energy waste, farmers can maintain optimal environmental conditions, leading to improved crop yields and reduced financial losses. In this study, multiple models, including Multiple Linear Regression (MLR), Radial Basis Function (RBF), and Support Vector Machine (SVM), were compared to predict greenhouse air temperature. External parameters, such as air temperature (Tout), relative humidity (Hout), wind speed (W), and solar radiation (S), were used as inputs for these models, and the output was the inside temperature. The results showed that the RBF model with the LM (Levenberg–Marquardt) learning algorithm outperformed the other models, achieving the lowest error and the highest coefficient of determination (R2) value. The RBF model produced RMSE, MAPE, and R2 values of 1.32 °C, 3.23%, and 0.931, respectively. These results demonstrate that the RBF model with the LM learning algorithm can reliably predict greenhouse air temperatures for the next two hours. The ANN model can be applied to optimize time management and reduce energy losses, improving the overall efficiency of greenhouse operations.

As consumer-grade depth sensors provide an efficient and low-cost way to obtain point cloud data, an increasing number of applications regarding the acquisition and processing of livestock point clouds have been proposed. Curve skeletons are abstract representations of 3D data, and they have great potential for the analysis and understanding of livestock point clouds. Articulated skeleton extraction has been extensively studied on 2D and 3D data. Nevertheless, robust and accurate skeleton extraction from point set sequences captured by consumer-grade depth cameras remains challenging since such data are often corrupted by substantial noise and outliers. Additionally, few approaches have been proposed to overcome this problem. In this paper, we present a novel curve skeleton extraction method for point clouds of four-legged animals. First, the 2D top view of the livestock was constructed using the concave hull algorithm. The livestock data were divided into the left and right sides along the bilateral symmetry plane of the livestock. Then, the corresponding 2D side views were constructed. Second, discrete skeleton evolution (DSE) was utilized to extract the skeletons from those 2D views. Finally, we divided the extracted skeletons into torso branches and leg branches. We translated each leg skeleton point to the border of the nearest banded point cluster and then moved it to the approximate centre of the leg. The torso skeleton points were calculated according to their positions on the side view and top view. Extensive experiments show that quality curve skeletons can be extracted from many livestock species. Additionally, we compared our method with representative skeleton extraction approaches, and the results show that our method performs better in avoiding topological errors caused by the shape characteristics of livestock. Furthermore, we demonstrated the effectiveness of our extracted skeleton in detecting frames containing pigs with correct postures from the point cloud stream.

In this paper, a deep-learning model is proposed as a viable approach to optimize the information on soil parameters and agricultural variables’ effect in cotton cultivation, even in the case of small datasets. In this study, soil is analyzed to reduce the planting costs by determining the various combinations of soil components and nutrients’ precise amounts. Such factors are essential for cotton cultivation, since their amounts are often not precisely defined, and especially traditional farming methods are characterized by excessive distribution volumes producing significant economic and environmental impact. Not only can artificial intelligence decrease the charges, but it also increases productivity and profits. For this purpose, a deep learning algorithm was selected among other machine learning algorithms by comparison based on the accuracy metric to build the predictive model. This model gets the combination of the factors amounts as input and predicts whether the cotton growth will be successful or not. The predictive model was built by this algorithm based on 13 physical and chemical factors has 98.8% accuracy.

Farm machinery selection, operation and management directly impact crop cultivation processes and outputs. A priori quantification of technical and financial needs allows definition of proportionate distribution and management of available resources and simplification of selection process. Appropriate planning, association and adjustment of the power unit and implement are required for soil cultivation. Consideration of functional parameters of the implement, their proper estimation and operation directly impact the soil structure, productivity and return on investment. Thus, a modelling approach was implemented for the definition of possible parameter-price relations for tillage equipment. The performed analysis allowed us to investigate the main relevant parameters, quantify their impact, and elaborate forecasting models for price, power, mass and working width. The significant relevance of the technical parameters and adjustment issues were outlined for each tillage implement group. For harrows and cultivators, the dependencies between studied parameters expressed better predictive qualities, especially for price-mass relation (R² > 0.8). While for ploughs power and mass relation had a primary output (R² = 0.7). The prediction features of the models provided reliable results for the estimation of the indicative values of the price and parameters of the implements.

Sensor applications are impacting the everyday objects that enhance human life quality. In this special issue, the main objective was to address recent advances of sensor applications in agriculture covering a wide range of topics in this field. A total of 14 articles were published in this special issue where nine of them were research articles, two review articles and two technical notes. The main topics were soil and plant sensing, farm management and post-harvest application. Soil-sensing topics include monitoring soil moisture content, drain pipes and topsoil movement during the harrowing process while plant-sensing topics include evaluating spray drift in vineyards, thermography applications for winter wheat and tree health assessment and remote-sensing applications as well. Furthermore, farm management contributions include food systems digitalization and using archived data from plowing operations, and one article in post-harvest application in sunflower seeds.

Technical and performance parameters of agricultural machines directly impact the operational efficiency and entire crop production. Sometimes, overestimation of technical and dimensional parameters of harvesting equipment is carried out with the intention of enhancing the operational efficiency, but this approach might turn out to negatively impact productivity due to unbalanced system design, and ultimately lead to financial losses. Therefore, a balanced preliminary estimation of technical parameters of equipment needs to be carried out before investment quantification, especially on the large capital-intensive machinery units, such as harvesting systems. In addition, availability of ready to use, simplified models for the price estimation from input technical parameters would reduce the complexity involved in this latter analysis. The current study is an attempt to provide tools to address these issues. A large dataset of combine and forage harvesters has been analyzed to investigate relevant parameter-to-parameter and parameter-to-price relations. The study of the available data allowed the determination of indicative models for the estimation of machine price, power, weight, tank capacity and working width. A significant correlation between power and price (R2 > 0.8) has been observed for two groups of harvesting machines. For combine harvesters, satisfactory correlations were found between power and weight, and power and tank capacity. A regression model for combine harvesters showed a satisfactory behavior at predicting the average working width that can be operated by a given power. On the other hand, for the forage harvesting group, the relation between these quantities has lower values; therefore, for better accuracy of the association, more sophisticated considerations should be incorporated, taking into account other parameters.

A correct soil moisture estimation is a fundamental prerequisite for many applications: agriculture, meteorological forecast, flood and drought prediction, and, in general, water accounting and management. Traditional methods typically provide point-like measurements, but suffer from soil heterogeneity, which can produce significant misinterpretation of the hydrological scenarios. In the last decade, cosmic-ray neutron sensing (CRNS) has emerged as a promising approach for the detection of soil moisture content. CRNS can average soil moisture over a large volume (up to tens of hectares) of terrain with only one probe, thus overcoming limitations arising from the heterogeneity of the soil. The present paper introduces the development of a new CRNS instrument designed for agricultural applications and based on an innovative neutron detector. The new instrument was applied and tested in two experimental fields located in Potsdam (DE, Germany) and Lagosanto (IT, Italy). The results highlight how the new detector could be a valid alternative and robust solution for the application of the CRNS technique for soil moisture measurements in agriculture.

Water scarcity is worsened by climate change. Water savings can be reached by improving irrigation efficiency both on farm and on water supply. To do that, the choice of the best irrigation technology is not always straightforward, because farmers need to renew and implement farm infrastructures for irrigation. This study compares three irrigation systems, one drip irrigation and two sprinkler (center pivot and hose-reel) systems, on environmental, economic, and energetic performance under irrigated and non-irrigated maize cropping. The study combines impact and efficiency indicators, addressing a sustainability analysis for the irrigation practice under the three different irrigation systems. The sustainability for the irrigation systems was assessed using water-related indicators (water use efficiency, irrigation water use efficiency, and water footprint), biomass (crop growth rate, relative growth rate, harvest index, and yield response factor), and energy indicators (energy footprint, performance, and energy cost footprint) for the environmental aspect; and the economic-based indicators (water productivity and economic water footprint) for the economic aspect. Main results address the center pivot system as the best solution for irrigation practice since it demonstrated higher economic and environmental performance. Moreover, maize under the pivot system allowed a higher biomass production, economic benefits, and water use efficiency.

Harrowing is a process that reduces the size of soil clods and prepares the field for seeding. Rotary harrows are a common piece of equipment in North Italy that consists of teeth rotating around a vertical axis with a processing depth of 5−15 cm. In this study, the topsoil movement in terms of distance and direction were estimated at different rotary harrow working conditions. A total of eight tests was performed using two forward speeds of 1 and 3 km/h, two working depths of 6 and 10 cm and two levelling bar positions of 0 and 10 cm from the ground. In order to simulate and follow topsoil movement, Radio-Frequency Identification (RFID) tags were inserted into cork stoppers and distributed in a regular pattern over the soil. Tags were distributed in six lines along the working width and repeated in three rows for each test: a total number of 144 tags was tracked. Results showed that there were no significant differences between the performed tests, on the other hand the reported tests highlight the effectiveness of the RFID monitoring approach.

Machine functional parameters define fleet composition and management and, thus, play an important role in economic and environmental performance. Large availability of programming methods and decision support systems are available in the market, however, there is still a lack of applicative tools to forecast the perceived and necessary technical parameters and machinery price options to complete tasks. In the current research, most correlated functional parameters for four group of seeding machines were determined with the application of linear and multiple linear regression analyses. Power, weight, working width, number of rows, and list price were studied, and reference equations were developed for seed drills, precision, combined and no-tillage planters. Two statistical analyses models were, therefore, developed for each of the groups in order to allow evaluation and prediction of performance and cost, thus contributing to the selection process optimisation and perceived choice of the needed implement.