Important Agricultural Products Research Articles

A Temporal–Geospatial Deep Learning Framework for Crop Yield Prediction

With the rapid development of information technology, the demand for digital agriculture is increasing. As an important agricultural production topic, crop yield has always attracted much attention. Currently, artificial intelligence, particularly machine learning, has become the leading approach for crop yield prediction. As a result, developing a machine learning method that accurately predicts crop yield has become one of the central challenges in digital agriculture. Unlike traditional regression prediction problems, crop yield prediction has a significant time correlation. For example, weather data for each county show strong temporal correlations. Moreover, geographic information from different regions also impacts crop yield to a certain extent. For example, if a county’s neighboring counties have a good harvest, then this county is likely to have high yields as well. This paper introduces a novel hybrid deep learning framework that combines convolutional neural network (CNN), graph attention network (GAT) and long short-term memory (LSTM) modules to enhance prediction accuracy. Specifically, CNN is employed to extract the features from the input data for each county in each year. GAT is introduced to model the geographical relationships between neighboring counties, allowing the model to capture spatial dependencies more effectively. LSTM is used to extract the temporal information within many years. The proposed hybrid deep learning framework CNN-GAT-LSTM captures both the temporal and spatial relationships, thereby improving the accuracy of yield prediction. We conduct experiments on a nationwide dataset that includes data from 1115 soybean-producing counties in 13 states in the United States covering the years from 1980 to 2018. We evaluate the performance of our proposed CNN-GAT-LSTM model based on three metrics, namely root of the mean squared error (RMSE), R-squared (R2) and correlation coefficient (Corr). The experimental results demonstrate that the proposed model achieves significant performance improvements over the existing state-of-the-art model, with RMSE reduced by 5%, R2 improved by 6% and Corr enhanced by 4%.

Landscape forest cover and local vegetation structure mediate multitrophic relationships but not the leaf damage in cacao trees

Cacao is the most important agricultural product in the southern region of Bahia state, Brazil, with 70 % of its production occurring under the traditional agroforestry where cacao is mostly shaded by native trees. This traditional system allows to reconcile the production with the maintenance of the portion of original biodiversity. However, increased deforestation and intensified agroforestry management aimed at boosting productivity may impact the diversity of native species and the services they provide. In this context, our aim was to disentangle the role of landscape forest cover and the local vegetation complexity on predation of caterpillars and herbivory of cacao plants located in agroforestry systems. The study was conducted across 18 cacao agroforest sites in southern Bahia located in landscapes with different amounts of forest cover. We assessed predation rate using dummy caterpillars, sampling understory birds and arthropods and collected leaves of cacao trees to analyze damage by herbivory. We also measured shading levels and the abundance of cacao trees in each agroforestry. Predation pressure on dummy caterpillars was positively influenced by the abundance of total predators and the level of landscape forest cover and negatively by the number of cacao trees. Even so, we found no evidence that landscape, local features or the actual invertebrate assemblages (predators or herbivores) influenced the cacao leaf damage. The findings highlight the multifaceted interactions between ecological factors, predation pressure, and leaf damage within cacao agroforestry systems.

Effect of D-mannitol and Gibberellic Acid on Physiological Parameters, Early Seedling Growth Characteristics, and Some of the Most Important Chemical Components of Barberry (Berberis vulgaris L.)

Background: One of the important medicinal plants and agricultural products in Asia, particularly in Iran, is barberry, and both salinity and drought can cause osmotic or ionic imbalance in plant cells. In order to study the impacts of D-Mannitol and gibberellic acid on seed germination, physiological parameters, early seedling growth characteristics, and some of the most important chemical ingredients of barberry, one experiment is designed. Methods: The experimental design was Completely Randomized Design (CRD) with three replicates, and drought stress was done by D-mannitol at various stress levels, namely, 0 MPa, 40 MPa, and 80 MPa, and the reaction of studied cultivar was assessed on the basis of experimental characteristics associated to germination and seedling growth, and Gibberellic acid treatments were 0 mgL-1, 300 mgL-1, 600 mgL-1, and 900 mgL-1. Results: The effects of D-mannitol were significant in germination percentage, germination rate, seedling vigor index, chlorophyll a, total chlorophyll, leaf dry weight and total sugar. Experimental treatments such as germination percentage, germination rate, and total chlorophyll were also influenced by the application of gibberellic acid. The interaction between D-mannitol and gibberellic acid had significant effect on peroxidase activity. The highest germination percentage, germination rate, seedling vigor index, catalase activity, peroxide activity, and Chlorophyll a were obtained for 0 MPa application of D-Mannitol, followed by usage of 40 MPa, and 80 Mpa, respectively. The maximum total chlorophyll, relative water content, and leaf dry weight were also related to control treatment (0 MPa D-Mannitol), while the higher values of proline content, carotenoids and total sugar were obtained for 80 MPa application of D-Mannitol. The maximum values of germination rate, germination percentage, seedling vigor index, peroxidase activity, and chlorophyll a were related to control treatment (0 mgL-1 gibberellic acid), and the highest CAT activity was related to the application of 300 mgL-1 gibberellic acid. The highest total chlorophyll content and relative water content were related to 0 mgL-1 gibberellic acid, while the maximum praline content and leaf dry weight were obtained for the application of 600 mgL-1 and 0 mgL-1 gibberellic acid. There was not any significant difference in carotenoids between gibberellic acid treatments, while the application of 900 mgL-1 gibberellic acid obtained the highest value of total sugar of barberry. Conclusion: Conclusively, on the basis of the results, the application of 80 MPa D-Mannitol and 600 mgL-1 could be recommended as they can increase the absorption of nutrients and water.

Evaluation of Enzymatic and Non-enzymatic Antioxidant Defense Responses of Durum Wheat (Triticum durum Desf.) in Coping with Boron Stress

Wheat, one of the world's most important agricultural products, plays a vital role in meeting the nutritional needs of our growing global population. However, arid and semi-arid regions face a potential threat from boron (B) toxicity. While boron is an essential nutrient for plant growth and development, its excessive presence can be toxic. It disrupts physiological processes, causing chlorosis and necrosis, ultimately leading to yield loss or plant death. Although B deficiency is a problem in the soils of many countries, Türkiye is one of the few experiencing B toxicity problems in its agricultural areas. This study investigated the physiological and biochemical responses of durum wheat to various B concentrations (0-20 mg L-1) under controlled air-conditioned cabin conditions. Durum wheat exhibited a decrease in chlorophyll content, phenolic content, and antiradical capacity at B doses exceeding 10 mg L-1. However, carotene content increased steadily with increasing B concentrations. The activities of antioxidant enzymes, including superoxide dismutase (SOD), ascorbate peroxidase (APX), and glutathione S-transferase (GST), increased up to a B dose of 15 mg L-1. Catalase (CAT) and glutathione reductase (GR) activities increased up to 10 mg L-1 B dose but decreased at higher B levels. Proline content increased tenfold up to a B dose of 10 mg L-1, indicating an attempt to mitigate stress. Conversely, malondialdehyde (MDA) accumulation increased continuously (approximately 150%) with increasing B doses, suggesting membrane damage. Despite being considered B-sensitive, this study demonstrated that durum wheat can effectively cope with B stress up to a B dose of 10 mg L-1 under controlled conditions. Beyond this threshold, physiological and biochemical changes indicate a decline in stress tolerance. Many osmoregulators, carotenes, alkaloids, flavonoids, tocopherols, phenolic compounds, non-protein amino acids, and several unidentified metabolites are activated to support antioxidant defense. The SOS pathway and the released ROS force gene regulatory systems come into play. Following these, the ROS released in the organism are neutralized, and ionic homeostasis and cellular stress resistance are achieved.

Occurrence of modified mycotoxins in Latin America: an up-to-date review.

The Latin America region has a considerable extent of varied climate conditions: from tropical, subtropical, and warm temperate to temperate. Among the surface territory, different agricultural products are produced, making them an important food source for human consumption. Fungal species commonly colonize those important agricultural products and often contaminate them with mycotoxins that have a major impact on health, welfare, and productivity. Nowadays, special attention is paid to modified mycotoxins, which are those that cannot be detected by conventional analytical methods. However, little data about their natural occurrence in food and feed is available, especially in Latin American countries, where, among all the countries in this region, only a few of them are working on this subject. Thus, the present review summarizes the published information available in order to determine the possible human exposure risk to these toxins.

Evolution and Drivers of Production Patterns of Major Crops in Jilin Province, China

Exploring the evolution of the production patterns of major crops and their driving factors can provide new ideas to ensure the security of supply of important agricultural products. This study selected four major crop types in Jilin Province—grain, oil crop, vegetable, and fruit—as research subjects. Using the center of gravity migration model, spatial autocorrelation analysis, and comparative advantage index, the spatial and temporal evolution of these crops’ production patterns from 2000 to 2020 was examined. A geographically weighted regression model was employed to analyze driving factors. The results showed that the planting area of grain was the most extensive; the percentage of production of grain, oil crop, and fruit increased to different degrees; the unit area yield of the four crops increased. The centers of gravity of all four crops’ productions were located in the central-northwestern region of Jilin Province. The global Moran’s I was positive for all four crops’ productions. The comparative advantage characteristics of the four crops were closely related to the natural geographical environment. Grain, vegetable, and fruit production were significantly and positively correlated with fertilizer application, and the most significant factor affecting oil crop production was land input intensity.

Promoting Peri-urban Agriculture: A Remedy for Climate Change. Case of the Town of Delly Brahim in Algiers

The issue of promoting peri-urban agriculture seems to be an important one in response tothe various current concerns about the future of cities in the face of climate change. At firstsight, these peri-urban agricultural areas, which in the past provided shelter for cities, arenow threatened by urban sprawl and are increasingly losing their function of protectingurban areas from climatic hazards. This question is a priority when considering the future ofagriculture in today's urban sprawl. We are therefore raising a number of concerns linked tomultiple issues: the trampling of green space, agricultural environmental sustainability,territorial impacts and the difficulty of managing rural and urban populations.Around these questions, this research is structured to provide theoretical and practicalknowledge that will enable us to place peri-urban agriculture at the heart of urban concerns,in a perspective of reaction to climate change. The interest of this work is reinforced in thecase of peri-urban areas surrounding the capital of Algiers, such as the commune of DelyBrahim: a concrete example of a territory concerned by the subject. On the one hand, thisperiphery boasts important agricultural production zones in close geographical proximity tothe capital Algiers, and on the other, it plays a fundamental economic, environmental andlandscape role in relation to its region. Between its dense urban residential areas, itsagricultural areas threatened by urban sprawl and its natural areas rich in resources, thisterritory is today confronted with numerous environmental vulnerabilities. The aim istherefore to examine the issue of preserving the agricultural sector and its role in mitigatingclimate change.

Productivity and technical inefficiency: A case study of small dryland mungbean farming in East Nusa Tenggara, Indonesia

Mungbeans are one of the important agricultural products for food security in East Nusa Tenggara (NTT) Province, which is one of the driest regions in Indonesia. The low productivity and profitability of mungbean farming in this region are caused by factors that influence agricultural inefficiency. This study was conducted to measure the productivity and technical efficiency levels, as well as investigate the factors affecting farm inefficiency, using the Technical Efficiency Effect Model (TEEM) applied to the frontier logarithmic transcendental production function. The method applied in this research is descriptive qualitative based on survey and interview results through primary and secondary data. Firstly, the results showed that mungbean farming was in the increasing return to scale stage of production and productivity only reached 57% of its production potential, with a technical efficiency level of 66%. Secondly, the factors that positively influence the low productivity and technical efficiency level of mungbean farming are land, seeds, labor, and the interaction between land and labor. Meanwhile, farmer experience and farm training significantly reduce the inefficiency of mungbean farming. It is suggested that increasing land productivity with appropriate and intensive modern technology and improving labor skills are the main strategies for reducing inefficiencies and increasing the income of smallholder farmers in drylands.

Data-Efficient Vision Transformer Models for Robust Classification of Sugarcane

Sugar cane is an important agricultural product that provides 75% of the world's sugar production. As with all plant species, any disease affecting sugarcane can significantly impact yields and planning. Diagnosing diseases in sugarcane leaves using traditional methods is slow, inefficient, and often lacking in accuracy. This study presents a deep learning-based approach for accurate diagnosis of diseases in sugarcane leaves. Specifically, training and evaluation were conducted on the publicly available Sugarcane Leaf Dataset using leading ViT (Vision Transformer) architectures such as DeiT3-Small and DeiT-Tiny. This dataset includes 11 different disease classes and a total of 6748 images. Additionally, these models were compared with popular CNN models. The findings of the study show that there is no direct relationship between model complexity, depth, and accuracy for the 11-class sugarcane dataset. Among the 12 models tested, the DeiT3-Small model showed the highest performance with 93.79% accuracy, 91.27% precision, and 90.96% F1-score. These results highlight that rapid, accurate, and automatic disease diagnosis systems developed using deep learning techniques can significantly improve sugarcane disease management and contribute to increased yields.

Experimental and simulation study on the performance of a solar assisted multi-source heat pump drying system in Zhengzhou area

Zhengzhou of Henan province is an important agricultural production base. To improve the performance of the combined drying system and promote the development of agricultural products in Zhengzhou area, a solar assisted multi-source heat pump (SMSHP) drying system was proposed based on a traditional solar assisted air source heat pump (SASHP) drying system. The simulation models were developed for the air source heat pump (ASHP), SASHP, and SMSHP drying system respectively. The coefficient of performance, energy consumption and specific moisture extraction rate of the drying systems under different seasons were analyzed. Moreover, to validate the simulation models, an experimental platform of the combined drying system was designed and set up, and the experimental tests were carried out on the ASHP, SASHP, and SMSHP drying system. The simulation results revealed that the SMSHP drying system had lower energy consumption, higher coefficient of performance and specific moisture extraction rate than that of the SASHP drying system. The most significant performance enhancement was observed in spring, where the SMSHP system demonstrated a 22.69 % increase in coefficient of performance, a 19.2 % increase in specific moisture extraction rate, and a 16.11 % decrease in energy consumption compared to the SASHP drying system. Comparing with the experimental data, the simulation model errors of the energy consumption, coefficient of performance, and specific moisture extraction rate of the SMSHP drying system were 3.19 %, 0.28 %, and 2.27 %, respectively. Therefore, the accuracy of the simulation results was confirmed. To better calculate the heat provided by the solar collection system, the concept of the tank heating guarantee rate was first introduced, and it was found to be more scientific than the solar energy guarantee rate. At last, The economic and environmental benefit of the drying systems were analyzed during the life cycle, and the SMSHP drying system exhibited better comprehensive performance than the SASHP drying system. This paper provides valuable insights into the comprehensive utilization of solar energy and the control logic of solar assisted air source heat pump combined drying system under different seasons.

Non-destructive Banana Quality Assessment and Quality and Safety Monitoring using Spectral Imaging Technology

This thesis discusses a method to realize non-destructive banana quality assessment and quality and safety monitoring using spectral imaging technology. As one of the important agricultural products in China, the quality and safety of bananas have always attracted much attention. Traditional quality assessment methods often require destroying bananas, but this method uses spectral imaging technology to realize the assessment of banana quality by measuring and analyzing the spectral characteristics of bananas. At the same time, this method also utilizes spectroscopic technology to detect harmful substances in bananas to realize the safety monitoring of banana quality. The experimental results show that the method has high accuracy and reliability, and can be used as a rapid, efficient, non-destructive means of banana quality assessment and quality and safety monitoring.

Morphological comparison of tympanal organs between Phycitinae and Pyralinae (Lepidoptera: Pyralidae)

Tympanal organ plays an important role for insects in perception, fleeing from predators, defense, finding and choosing sex, communication, and carrying out social interactions. It is positioned in different areas among various insect groups. Tympanal organ is in the abdomen in the Pyralidae family, many species of which are harmful to economically important agricultural products and storage products. This study was conducted to determine the importance of the tympanal organ in the classification of Pyralids. For this purpose, tympanal organs of 10 species (Pyralis farinalis (Linnaeus, 1758), Hypsopygia glaucinalis (Linnaeus, 1758), Endotrichia flammealis ([Denis & Schiffermüller], 1775), Dioryctria pineae (Staudinger, 1859), D. mendacella (Staudinger, 1859), Oncocera semirubella (Scopoli, 1763), Etiella zinckenella (Treitschke,1832), Acrobasis obliqua (Zeller, 1847), Myelois circumvoluta (Geoffroy in Fourcroy, 1785), Ematheudes punctellus (Treitschke, 1833) in four tribus belonging to Pyralinae and Phycitinae were examined for the first time. Bulla tympani, fornix tympani, pons tympani, rugae odontinae, intersegmental thoraco abdominal membrane, processus spiniforme etc. structures of these species were described and photographed for the first time. The similarities and differences between taxa were analyzed.

EXPERIMENTAL STUDY ON THE EFFICACY OF ULTRASOUND IN IN-SITU DETECTION OF CORROSION IN PALM OIL REFINING EQUIPMENT

Palm oil is an important agricultural product for many countries. Stainless steel refining equipment faces corrosion challenges due to the corrosive nature of palm oil, which gradually deteriorates the equipment over time. Ultrasonic testing (UT) can detect corrosion, usually performed during palm mill downtime due to temperature constrain. The study aims to optimize UT at temperatures up to 250°C. The experimental outcomes serve as a foundation for broadening the application of phased array corrosion mapping (PACM) into elevated temperatures. The results offer intricate insights into velocity fluctuations within specific materials, correlating with temperature elevation. In summary, the suite of UT methods demonstrates their aptitude for employment during palm oil refinery operations up to 250°C involving carbon steel as well as 304 and 316 stainless steels. It allows further UT application during the palm mill’s operation and achieves the study’s objective of allowing palm mill owners to detect and monitor corrosion growths without stopping production.

Bio-stimulant based nanodelivery system for pesticides with high adhesion and growth stimulation

Pesticides are important agricultural production materials to ensure global food security, but the low utilization rate during pesticide application has caused a large amount of waste and environmental pollution. The use of pesticide carriers to deliver pesticides can help to prevent rapid release, assist in targeted release, and improve pesticide utilization. However, after completing pesticide delivery, less part of pesticide carriers has more functions and may remain in the environment, leading to material waste and environmental burden. Here, we propose that the environment-friendly carrier materials for pesticide delivery can synergistically enhance control efficacy with pesticides. Based on the bio-stimulant polyglutamic acid, amphiphilic polymers (mPEG-b-PLG) were constructed, and they assembled into nano micelle. After encapsulating prothioconazole (PTC) into the nano micelle through hydrogen bonding, electrostatic interactions and hydrophobic interactions through co-assembly, mPEG-b-PLG-PTC nanospheres with a diameter of approximately 120 nm were prepared. The nanospheres exhibit pH responsiveness for targeted release of pesticides, as well as excellent retention capacity, droplet bounce inhibition ability, and deposition capacity. At the same time, compared with commercial formulations, it has long-term antifungal properties and lower EC50, while its acute toxicity to human cells is lower, which can reduce cell apoptosis by about 63 %. It is also gratifying that the main root length, plant height, and fresh weight of wheat seedlings treated with mPEG-b-PLG have significantly improved, effectively applying carrier materials for synergistic effects. In summary, this work proposes a promising strategy to synergize pesticide carrier materials as active ingredients, providing new ideas for the integration of pesticide and fertilizer, improving pesticide utilization efficiency, and reducing environmental pollution.

Illuminating the path to sustainable rice harvesting: A solar-based paddy harvester

Paddy is the most important agricultural crop product in Bangladesh since it is widely used in daily life as a main meal intake. Currently, traditional manual harvesting process is widely used to harvest paddy grains in Bangladesh. However, manual harvesting is laborious, time-consuming, and costly. Utilization of modern machineries such as combined paddy harvester machines can increase grain productivity while reducing significantly the number of labour-hours. Combine rice harvesters possess traditional diesel engine that burns gasoline fuels i.e. diesel. The overriding concern of this diesel oil combustion is the production of greenhouse gases that adversely affect the environment. In addition, reports suggest that oil reserve might be depleted to zero in next 40 years. This manuscript presents a comprehensive review on the paddy grain production statistics, the pros and cons of traditional rice crop harvesting processes, and the historical development of mechanization in the agricultural sectors of Bangladesh. It also presents a novel design route to convert commercial diesel-driven combine paddy harvester machine to solar energy-operated combine rice harvester machine with its required accessories. In particular, a new design concept with a remarkable cost estimation process of solar-powered harvesters is proposed. A field study has been conducted in Sonagazi Upazilla under Feni district of Bangladesh to evaluate the potential use of the proposed solar-powered harvester machine. Our study suggests that solar-driven rice harvester is a promising concept in the paddy crop harvesting process and it will reduce the overall costs for short and long-term uses compared to estimated costs of commercial diesel-driven harvesters. To combat the scarcity of gasoline in the future, it is an obvious choice to bring a paradigm shift in solar energy harvesting and its utilization in the agricultural sectors of Bangladesh. It is expected that the proposed solar-driven rice harvester concepts with their in-depth analysis (including costs) would be able to draw attention to business entrepreneurs and investors who will be excited to initiate a program to manufacture solar-powered rice harvester machines in practice.

Use of the HEC RAS model for the analysis of exceptional floods in the Ouémé basin

Abstract. The Ouémé River basin extends over almost half of Benin's territory, entirely located in a humid tropical climate. This river system includes a deltaic zone (delta of the Ouémé) known for its high agricultural potential and thus subject to a socio-economic development agenda. The Ouémé delta is facing recurrent floods that maintain rural agricultural population into a retrograding crisis with significant damages such as losses of properties. The objective of this study is to improve decision-making in the Ouémé basin through the simulation of exceptional floods using the HEC-RAS model. The HEC RAS model is a conceptual model, which works through mathematical and physical formulas to implement environmental phenomena for forecasting, understanding and analysis purposes. The model inputs used are basin GIS data, hydro-meteorological data, characteristics of existing hydraulic structures, etc. The targeted outputs include 1D/2D/3D view plans with support of satellite images, tables, graphs and curves. It is worth mentioning that the model provides outputs compatible with other tools, such as civil engineering (Civil 3D, Revit, Infraworks, etc.) and GIS, that help to expand the valorization fields. The implementation of the model in the Ouémé basin has made it possible to note: (i) that the recurring effect of losses and damages is justified by the settlement of the population on the river banks; (ii) that there is an important agricultural production in areas of high flood risk; (iii) that depending on the occurrence of the phenomenon, the flooded extent and the height of submersion remains variable, and more important for extreme flooding; (iv) about 12.07 % occurrence of river flood against 13.24 % for flash flood at a return period of 30 years. Moreover, it is very relevant to note that most of flood waters converge to the western part of the basin (an area with a low risk of flooding, stretched over 63.68 km2) and to the eastern part around the Damè-Wogon depression (an area at high risk of flooding, stretched over 10.49 km2).

Safety of agricultural machinery and tractor maintenance planning with fuzzy logic and MCDM for agricultural productivity

Productivity is one of the most important measures used to determine the growth and development level of countries or sectors. A wide variety of projects have been planned and implemented to increase agricultural productivity. The productivity to be obtained in agriculture; Soil conditions, climate, seeds, fertilizer, pesticides, labor and agricultural mechanization directly affect it. Agricultural mechanization is the realization of agricultural activities by using energy together with agricultural tools and machines. Agricultural mechanization; It is an important agricultural production technology that helps increase agricultural productivity. Due to the inadequate maintenance planning of agricultural machinery, agricultural machinery cannot be utilized at the desired level in agricultural production. Most agricultural equipment is subject to frequent changes in speed and direction of movement while operating. Damage that can be seen on a single machine; It also causes other machines to malfunction. During the year, especially in the months when agricultural activity is high, excessive working tempo can cause tractors to malfunction. The breakdown of tractors causes disruptions in agricultural activities. In addition, the breakdown of tractors increases the repair costs. Since there is no tractor maintenance planning, farmers face interruptions in agricultural activities due to tractor malfunction. However, tractor malfunctions may cause cost and economic losses. For these reasons, there is a need for appropriate maintenance planning of agricultural machinery in order to continue agricultural activities without disruption. Maintenance planning; It consists of a set of preventive activities to improve the reliability and availability of any system. The main purpose of this study is to determine and rank the importance level weights of the criteria that are important for agricultural machinery maintenance planning using the fuzzy AHP method. Fuzzy AHP method, which provides ease of application, was preferred in determining the Criterion Weights. The research proposes a framework to determine the weights of appropriate criteria for care planning selection through a combined approach of fuzzy multi-criteria decision making involving relevant stakeholders. On the basis of the prioritization of criteria of tractor maintenance planning (TMP), it was found from the ranking that checking for all fluid levels (TMP1) ranked first. This respectively is followed by checking for general conditions (TMP4), checking for tires and wheels (TMP2) and checking for batteries (TMP3). With the results of the study, a guide was created for farmers and other stakeholders, as well as decision makers, to help plan the maintenance of machines in better working conditions. It is also thought that this study will be encouraging for other studies.

TeaViTNet: Tea Disease and Pest Detection Model Based on Fused Multiscale Attention

The tea industry, as one of the most globally important agricultural products, is characterized by pests and diseases that pose a serious threat to yield and quality. These diseases and pests often present different scales and morphologies, and some pest and disease target sizes can be tiny and difficult to detect. To solve these problems, we propose TeaViTNet, a multi-scale attention-based tea pest and disease detection model that combines CNNs and Transformers. First, MobileViT is used as the feature extraction backbone network. MobileViT captures and analyzes the tiny pest and disease features in the image via a self-attention mechanism and global feature extraction. Second, the EMA-PANet network is introduced to optimize the model’s learning and attention to the Apolygus lucorum and leaf blight regions via an efficient multi-scale attention module with cross-space learning, which improves the model’s ability to understand multi-scale information. In addition, RFBNet is embedded in the module to further expand the perceptual range and effectively capture the information of tiny features in tea leaf images. Finally, the ODCSPLayer convolutional block is introduced, aiming to focus on acquiring richer gradient flow information. The experimental results show that the TeaViTNet model proposed in this paper has an average accuracy of 89.1%, which is a significant improvement over the baseline network MobileViT and is capable of accurately detecting Apolygus lucorum and leaf blight of different scales and complexities.

An Apple Detection and Localization Method for Automated Harvesting under Adverse Light Conditions

The use of automation technology in agriculture has become particularly important as global agriculture is challenged by labor shortages and efficiency gains. The automated process for harvesting apples, an important agricultural product, relies on efficient and accurate detection and localization technology to ensure the quality and quantity of production. Adverse lighting conditions can significantly reduce the accuracy of fruit detection and localization in automated apple harvesting. Based on deep-learning techniques, this study aims to develop an accurate fruit detection and localization method under adverse light conditions. This paper explores the LE-YOLO model for accurate and robust apple detection and localization. The traditional YOLOv5 network was enhanced by adding an image enhancement module and an attention mechanism. Additionally, the loss function was improved to enhance detection performance. Secondly, the enhanced network was integrated with a binocular camera to achieve precise apple localization even under adverse lighting conditions. This was accomplished by calculating the 3D coordinates of feature points using the binocular localization principle. Finally, detection and localization experiments were conducted on the established dataset of apples under adverse lighting conditions. The experimental results indicate that LE-YOLO achieves higher accuracy in detection and localization compared to other target detection models. This demonstrates that LE-YOLO is more competitive in apple detection and localization under adverse light conditions. Compared to traditional manual and general automated harvesting, our method enables automated work under various adverse light conditions, significantly improving harvesting efficiency, reducing labor costs, and providing a feasible solution for automation in the field of apple harvesting.

Impacts of Crop Type and Climate Changes on Agricultural Water Dynamics in Northeast China from 2000 to 2020

Northeast China (NEC) is one of the most important national agricultural production bases, and its agricultural water dynamics are essential for food security and sustainable agricultural development. However, the dynamics of long-term annual crop-specific agricultural water and its crop type and climate impacts remain largely unknown, compromising water-saving practices and water-efficiency agricultural management in this vital area. Thus, this study used multi-source data of the crop type, climate factors, and the digital elevation model (DEM), and multiple digital agriculture technologies of remote sensing (RS), the geographic information system (GIS), the Soil Conservation Service of the United States Department of Agriculture (USDA-SCS) model, the Food and Agriculture Organization of the United Nations Penman–Monteith (FAO P-M) model, and the water supply–demand index (M) to map the annual spatiotemporal distribution of effective precipitation (Pe), crop water requirement (ETc), irrigation water requirement (IWR), and the supply–demand situation in the NEC from 2000 to 2020. The study further analyzed the impacts of the crop type and climate changes on agricultural water dynamics and revealed the reasons and policy implications for their spatiotemporal heterogeneity. The results indicated that the annual average Pe, ETc, IWR, and M increased by 1.56%/a, 0.74%/a, 0.42%/a, and 0.83%/a in the NEC, respectively. Crop-specifically, the annual average Pe increased by 1.15%/a, 2.04%/a, and 2.09%/a, ETc decreased by 0.46%/a, 0.79%/a, and 0.89%/a, IWR decreased by 1.03%/a, 1.32%/a, and 3.42%/a, and M increased by 1.48%/a, 2.67%/a, and 2.87%/a for maize, rice, and soybean, respectively. Although the ETc and IWR for all crops decreased, regional averages still increased due to the expansion of water-intensive maize and rice. The crop type and climate changes jointly influenced agricultural water dynamics. Crop type transfer contributed 39.28% and 41.25% of the total IWR increase, and the remaining 60.72% and 58.75% were caused by cropland expansion in the NEC from 2000 to 2010 and 2010 to 2020, respectively. ETc and IWR increased with increasing temperature and solar radiation, and increasing precipitation led to decreasing IWR in the NEC. The adjustment of crop planting structure and the implementation of water-saving practices need to comprehensively consider the spatiotemporally heterogeneous impacts of crop and climate changes on agricultural water dynamics. The findings of this study can aid RS-GIS-based agricultural water simulations and applications and support the scientific basis for agricultural water management and sustainable agricultural development.