Agriculture, Volume 14, Issue 4 (April 2024) – 138 articles

During the process of walnut identification and counting using UAVs in hilly areas, the complex lighting conditions on the surface of walnuts somewhat affect the detection effectiveness of deep learning models. To address this issue, we proposed a lightweight walnut small object recognition method called w-YOLO. We reconstructed the feature extraction network and feature fusion network of the model to reduce the volume and complexity of the model. Additionally, to improve the recognition accuracy of walnut objects under complex lighting conditions, we adopted an attention mechanism detection layer and redesigned a set of detection heads more suitable for walnut small objects. A series of experiments showed that when identifying walnut objects in UAV remote sensing images, w-YOLO outperforms other mainstream object detection models, achieving a mean Average Precision (mAP0.5) of 97% and an F1-score of 92%, with parameters reduced by 52.3% compared to the YOLOv8s model. Effectively addressed the identification of walnut targets in Yunnan, China, under the influence of complex lighting conditions. Full article

Forty wheat samples of ten wheat varieties harvested from optimal or late sowings in 2019 and 2020 were evaluated for nutrient composition. This included crude protein (CP), starch, amino acids, minerals, phytate-phosphorus (phytate-P) and non-starch polysaccharides (NSPs). The objective was to investigate the impact of high temperature on wheat grain quality as a feedstuff for broiler chickens. Growth performance and economic impact of such changes were predicted by the Emmans, Fisher and Gous broiler growth model. On average, 2019 was 1 °C hotter than 2020 during the growing season (Narrabri, NSW 2390, Australia). The wheat harvested in 2019 had higher concentrations of CP, phytate-P, total P and calcium. In 2019, late sowing increased average protein concentrations from 166.6 to 190.2 g/kg, decreased starch concentration from 726 to 708 g/kg and increased total NSPs from 693 to 73.9 g/kg. Unlike the 2019 harvest, the late sowing in 2020 had no impact on CP concentrations in almost all wheat varieties. The 2019 varieties had higher concentrations of 16 assessed amino acids (p < 0.001) compared to the 2020 harvest. The largest difference was in lysine (19.2%), and the smallest difference was in proline (11.1%). It was predicted that broiler diets formulated from 2019 wheat varieties would have better efficiency of feed conversion with an advantage of 2.53% (1.539 versus 1.579) than 2020 varieties to 35 days post-hatch. This would translate to a cost saving of approximately AUD 16.45 per tonne of feed, much of which would represent additional profit. Full article

The study examines the dichotomy between individual dietary autonomy and the broader implications of food overconsumption and waste, particularly focusing on meat consumption’s environmental, health, and social equity aspects. In the context of increasing awareness about the negative impacts of excessive meat consumption, this research explores the potential benefits of modest dietary shifts, specifically a reduction in animal product intake, on natural resources and the environment. Utilizing data from international and Romanian sources, including data about meat environmental impacts, in original research, the article analyzes the water, carbon, and land use footprints associated with different types of meat, emphasizing the significant differences between beef, pork, chicken, and sheep meat. The findings highlight that even a small reduction in meat consumption, such as 100 g per week per capita, can lead to substantial decreases in water use, carbon emissions, and land use, underscoring the importance of sustainable eating habits. Moreover, the study explores the potential of plant-based proteins as viable nutritional alternatives that can mitigate environmental footprints and foster global food security. Conclusively, this work advocates for a balanced approach that respects individual choices while promoting collective responsibility towards sustainable consumption patterns, emphasizing the role of scientific research and public awareness in driving positive change in dietary habits for environmental conservation and health benefits. Full article

Soil degradation threatens global food security and environmental sustainability, necessitating effective soil management strategies. This review comprehensively examines the impact of organic soil amendments on soil quality and productivity across various soil types and climatic conditions. A review of significant research related to organic amendments was performed using encompassed data from online search engines for studies published up until 31 December 2023. Despite their heterogeneity and use of varying methodologies, the data were narratively synthesized, providing a comprehensive understanding of amendment-induced changes in the chemical and physical properties of soil and the effectiveness of restoration on soil degradation. Organic amendments, including compost, vermicompost, biochar, and pomace, are pivotal in enhancing soil quality by increasing soil organic matter content, fostering aggregate formation, and improving soil structure in the short term. They positively influence water retention capacity, pH levels, nutrient availability, and carbon sequestration. In several studies, amendment-induced changes were absent, indicating that the effects of amendments vary depending on soil texture, application rates, and cropping systems, which emphasizes the need for tailored, sustainable soil management practices. This study concludes that organic amendments are a promising option for structure improvement and organic matter accumulation. It further suggests that an approach that integrates various methods is essential in order to meet desirable soil quality and retain agricultural productivity and offers valuable insights and recommendations for policymakers, practitioners, and researchers. Organic amendments can improve soil ecosystem services and contribute to climate change adaptation. In the future, more attention should be directed to tillage management and soil amendment interaction, as well as their effectiveness over specific periods of time. Full article

Plant factories with artificial lighting (PFALs) are indoor crop production systems aiming at the growth of high-value products in terms of yield and quality, while maximizing resource use efficiency. The emergence of PFALs opened a new world for crop production and offered an option to tackle problems related to climate change, land availability, and urban/peri-urban farming. This was made possible upon major technological advancements and extensive research in the field of controlled environment agriculture, which paved the way for the establishment of such cost-efficient and climate-unaffected modules of vegetable and other crops’ production. In the present review, we have examined the recent research achievements regarding the micro-environmental factors, the principal components, as well as the automated systems used for plant production in PFALs. Ultimately, we provide the reader with a number of future perspectives that can be considered for indoors cultivation in the following years. Full article

Background: With the purpose of exploring the development of new quality productive forces in Guangdong, the present study hypothesized that reducing energy and carbon inputs was beneficial for increasing Guangdong’s multi-cropping agricultural energy output and economic returns. Methods: The energy use efficiency of crop production and the corresponding carbon input and agricultural benefit trends from 2011 to 2021 were examined by linear regression analysis for Guangdong Province, China. The corresponding development trends were also predicted using a grey model. Results: The results indicated that the total energy output increased by 12.50%, wherein the energy productivity levels of peanuts, vegetables, mulberry, and fruits increased greatly (51.27–106.17%), while the total energy input and the total carbon input decreased by 12.17% and 6.37%, respectively. Moreover, the energy input per carbon input decreased by 6.19%, while the energy output per carbon input increased by 20.15%. Both energy-related indicators and economic-related indicators all had substantially increased (28.08–44.97% and 83.86–120.91%, respectively). Grey model predictions show that the agricultural output value increased steadily under the current agricultural policy of reducing fossil energy input. Conclusions: The current low-carbon and high-output agricultural model is beneficial for increasing Guangdong’s multi-cropping agricultural economic returns and mitigating greenhouse effects. Full article

The aim of this research work was to find historical varieties that existed in this area before phylloxera and to identify them on the basis of historical written sources and genetic analyses. At the beginning of the 19th century, around 300 grape varieties were cultivated in Styria. Between 2020 and 2022, old vineyards were monitored at 115 locations in Styria (between the Mura and Sava rivers) in Slovenia. The directly collected samples (340 grapevine accessions) were determined by molecular analysis with 24 SSR markers. A total of 66 different genotypes were detected. After comparison with the available databases, 29 historical varieties and 37 unknown historical genotypes were identified. Several parameters were calculated to evaluate the usefulness of the selected loci in this work, and a dendrogram representing the genetic similarities between the origins was created using the neighbor-joining method to investigate possible ancestry relationships in the sample set. The most common historical varieties were ‘Belina’ (‘Heunisch weiss’), ‘Vrbovec’ (‘Tantovina Eihenblaetrige’), ‘Ranfol’ (‘Ranfol beli’), and ‘Pelesovna’ (‘Vulpea’). Varieties from the current variety list were also frequently found, such as ‘Frankinja’ (‘Blaufraenkisch’) and ‘Žametovka’ (‘Kavčina črna’). In a few locations, one of the most important red varieties from the beginning of the 20th century was also found in this area (alongside ‘Frankinja’ and ‘Žametovka’), i.e., ‘Vranek’ (‘Zimmettraube balu’). At that time, this variety was planted in multi-variety vineyards and was preserved, but its importance in single-variety vineyards quickly declined due to female flower. In addition, genetic analyses have shown that 37 unknown historical genotypes have been found in this area. These genotypes need to be described ampelographically and technologically evaluated in the future. Most of the vegetative offsprings of these genotypes have already been transferred to the Meranova gene bank, where they can be accurately described ampelographically under the same pedoclimatic conditions. Full article

The solid-state fermentation (SSF) efficiency of Bacillus licheniformis ATCC 21424 (BL) on various agro-industrial by-products such as oilseed cakes [hemp (HSC), pumpkin (PSC), and flaxseed (FSC)] was evaluated by examining the nutritional composition, reducing sugars, and in vitro protein digestibility (IVPD) for use in animal nutrition. SSF significantly decreased crude protein, along with changes in the total carbohydrates (p < 0.05) for all substrates fermented. An increase in crude fat for HSC (1.04%) and FSC (1.73%) was noted, vs. PSC, where the crude fat level was reduced (−3.53%). Crude fiber does not differ significantly between fermented and nonfermented oilseed cakes (p > 0.05). After fermentation, neutral detergent fiber (NDF) and acid detergent fiber (ADF) significantly increased for HSC and FSC (p < 0.05), as well as for PSC despite the small increase in ADF (4.46%), with a notable decrease in NDF (−10.25%). During fermentation, pH shifted toward alkalinity, and after drying, returned to its initial levels for all oilseed cakes with the exception of PSC, which maintained a slight elevation. Further, SSF with BL under optimized conditions (72 h) increases the reducing sugar content for FSC (to 1.46%) and PSC (to 0.89%), compared with HSC, where a reduction in sugar consumption was noted (from 1.09% to 0.55%). The viable cell number reached maximum in the first 24 h, followed by a slowly declining phase until the end of fermentation (72 h), accompanied by an increase in sporulation and spore production. After 72 h, a significant improvement in water protein solubility for HSC and FSC was observed (p < 0.05). The peptide content (mg/g) for oilseed cakes fermented was improved (p < 0.05). Through gastro-intestinal simulation, the bacterial survivability rate accounted for 90.2%, 101.5%, and 85.72% for HSC, PSC, and FSC. Additionally, IVPD showed significant improvements compared to untreated samples, reaching levels of up to 65.67%, 58.94%, and 80.16% for HSC, PSC, and FSC, respectively. This research demonstrates the advantages of oilseed cake bioprocessing by SSF as an effective approach in yielding valuable products with probiotic and nutritional properties suitable for incorporation into animal feed. Full article

Effective estimation of crop yields at a regional scale holds significant importance in facilitating decision-making within the agricultural sector, thereby ensuring grain security. However, traditional ground-based measurement techniques suffer from inefficiencies, and there exists a need for a reliable, precise, and effective method for estimating regional rice yields. In this study, we employed four machine-learning techniques: partial least squares regression (PLSR), support vector regression (SVR), random forest regression (RFR), and back propagation neural network (BPNN). We combined these methods with multi-temporal rice NDVI (normalized difference vegetation index) data for rice yield estimation. Following an accuracy evaluation and a spatial analysis, the key findings of our study are as follows. (1) The RFR model emerged as the most accurate for rice yield estimation, achieving an R² of 0.65, an RMSE of 388.79 kg/ha, and an rRMSE of 4.48%. While PLSR and SVR demonstrated comparable accuracy, they were both inferior to RFR. (2) Using the top seven predictors with the highest importance rankings as inputs for the RFR model (NDVI values on the 6th, 17th, 33rd, 44th, 71st, 90th, and 106th days after the rice transplanting stage) achieved comparable accuracy while reducing information redundancy. (3) The proposed model demonstrated good spatial applicability (MI = −0.03) for rice yield estimation in Jiangsu, China. (4) A high spatial resolution yearly rice yield dataset (1 km) spanning from 2001 to 2020 was generated using the proposed model and is accessible on the Zenodo database. In conclusion, this study has demonstrated the efficacy of combining multi-temporal remote sensing data with machine-learning techniques for accurate rice yield estimation, thereby aiding agricultural authorities and production enterprises in the timely formulation and refinement of cropping strategies and management policies for the ongoing season. Full article

In order to solve the problem of image boundary segmentation caused by the irregularity of paddy fields in southern China, a high-precision segmentation method based on the improved MultiResUNet model for paddy field mapping is proposed, combining the characteristics of paddy field scenes. We introduce the attention gate (AG) mechanism at the end of the encoder–decoder skip connections in the MultiResUNet model to generate the weights and highlight the response of the field ridge area, add an atrous spatial pyramid pooling (ASPP) module after the end of the encoder down-sampling, use an appropriate combination of expansion rates to improve the identification of small-scale edge details, use 1 × 1 convolution to improve the range of the sensory field after bilinear interpolation to increase the segmentation accuracy, and, thus, construct the AM-UNet paddy field ridge segmentation model. The experimental results show that the IoU, precision, and F1 value of the AM-UNet model are 88.74%, 93.45%, and 93.95%, respectively, and that inference time for a single image is 168ms, enabling accurate and real-time segmentation of field ridges in a complex paddy field environment. Thus, the AM-UNet model can provide technical support for the development of vision-based automatic navigation systems for agricultural machines. Full article

Cultural crop-production practices are not only engineered to minimize pest incidence but also improve resource use efficiency and increase the diversity of habitat for beneficial insects that provide pollination services. With the increasing cultivation of industrial hemp and the benefits associated with the cultivation of multiple crops, its integration into a polyculture cropping system remains to be evaluated. We intercropped two pollinator-attractive crops, hemp and cowpea, with squash, a pollinator-dependent crop, to evaluate the impact of pollinator abundance and diversity on crop yield. Intercropping significantly increased the overall abundance of pollinators with 79.1% recorded from the intercropping systems compared to 21.9% in the monocropping systems. Sweat bees and bumble bees were the most abundant bees, and Squash+Cowpea was the most diverse cropping system. Intercropping significantly increased the yield of squash with higher squash yield (155%) in Hemp+Squash and (161%) in Squash+Cowpea than in squash monocrop. Also, intercropping resulted in higher hemp yield (64%) in Hemp+Cowpea and (165%) in Hemp+Squash compared to hemp monocrop. This study demonstrated that agricultural systems such as intercropping that are designed to attract pollinators are much more productive by not only improving crop yield but also growers’ returns on investments. Full article

Fertilizer can improve the yield of crops per unit area, and uniform fertilizer discharge can improve the fertilizer utilization rate. Therefore, it is meaningful to improve the performance of fertilizer-discharge devices in order to improve the modernization level of crop field fertilizer management. To address the problems of operational smoothness, stability and poor uniformity of fertilizer discharge, and other difficult problems encountered with strip fertilizer-discharge devices, this study designs a disturbed fertilizer strip-discharge device with a vertical pendulum. The main factors affecting the performance of fertilizer discharge were the wedge angle of the push-disturbing main pendulum bar (PMPB), the inclination angle of the aided-stirring pendulum pick (APP), the flow gap of the pendulum bar (FGPB), and the operation frequency of the swing-rod combination (SRC). The discrete element method (DEM) was used to establish a simulation model of the fertilizer device to explore the influence of the main factors on the performance of fertilizer discharge, with the coefficient of variation (CV) of fertilizer discharge uniformity and fertilizer discharge accuracy (FDA) used as the evaluation indices. The results show that the factors affecting the CV of fertilizer discharge uniformity and FDA were, in order of priority, the operation frequency of the SRC, the FGPB, the wedge angle of the PMPB, and the inclination angle of the APP. The optimal parameters after rounding were as follows: the wedge angle of the PMPB was 45°, the inclination angle of the APP was 46°, the operation frequency of the SRC was 188 times/min, and the FGPB was 4.5 mm. At this point, the model predicted that the CV of fertilizer discharge uniformity would be 10.53%, and that the FDA would be 3.19%. Using the optimal parameters for bench test verification, it was found that the wedge angle of the PMPB was 45°, the inclination angle of the APP was 46°, the operation frequency of the SRC was 188 times/min, the FGPB was 4.5 mm, the CV of the uniformity of the fertilizer discharge was 11.06%, and the FDA was 3.51%. In the test, the fertilizer-discharge device was stable and had good adaptability to different fertilizers. The results of this study can provide a theoretical reference for the development of precision strip-fertilizer application devices. Full article

To explore the effects of particle size distribution on its physicochemical, functional, and structural properties, alfalfa leaf powders with mean particle sizes (D₅₀) of 506.1, 246.3, 209.8, 92.01, and 20.68 μm were prepared by sieving. The physicochemical, functional, and structural properties of alfalfa were compared, and correlation and principal component analyses were conducted. As the D₅₀ of alfalfa leaf decreased, the bulk density, tap density, and the swelling capacity increased first and then decreased, but the compressibility, transition temperature, and melting temperature exhibited an opposite trend. The solubility, lightness, and inhibition of angiotensin-converting enzymes and tyrosinase were enhanced. Specifically, the alfalfa leaf with a D₅₀ of 209.8 μm exhibited a higher bulk density and swelling capacity and a lower compressibility, transition temperature, and melting temperature. The alfalfa leaf with a D₅₀ of 20.68 μm presented better solubility, lightness, and inhibition of angiotensin-converting enzymes and tyrosinase. Additionally, the surface roughness and the number of surface hydroxyls improved and the crystallinity index decreased, but the type of surface functional groups was unchanged. These changes in microstructure can provide an explanation for the trend of the physicochemical and functional properties. Moreover, based on the results of the correlation analysis and principal component analysis, it can be concluded that there are strong correlations among the particle size, physicochemical properties, and functional properties of alfalfa leaf. Overall, this conclusion can help determine the appropriate grinding particle size range for alfalfa leaf in different functional food products. Full article

Grain quality, cooking quality, and nutritional quality traits are some of the major attributes that enhance the uptake and utilization of improved cowpea varieties. Therefore, there is a need for a better understanding of the genetic variation and inter-relationships among these quality traits in cowpeas to integrate them into cowpea breeding programs. This study was conducted to determine genetic variability among 306 cowpea genotypes for grain quality, cooking quality, and nutritional quality traits and to understand the interrelationships among these traits for exploitation in breeding programs. The results showed highly significant differences (p < 0.001) among genotypes for grain quality, cooking quality, and nutritional quality traits. The mean performance for these quality traits was also very variable. These results suggest that genetic variability exists in the cowpea genotypes studied, which can be exploited in breeding programs aimed at developing high-performing varieties for the said traits. Significant (p < 0.001) positive correlations were detected for protein content with iron and zinc. On the other hand, nutritional quality traits did not exhibit any association with grain quality or cooking quality traits. Cooking quality traits were also shown to be significantly and positively correlated with grain quality traits. This study has identified several genotypes with desirable quality-related traits that could be used in crossing programs to generate improved varieties with consumer-preferred traits to improve the food, income, and nutritional status of many smallholder farmers that largely depend on cowpeas. Full article

With the improvement in living standards and growing appreciation for flaxseed’s nutritional value, global demand for flaxseed and its economic significance are continuously increasing. As a major flaxseed producer and exporter, China plays a crucial role in the development of its agricultural economy. Flaxseed, one of China’s five key oil crops, is renowned for its rich nutritional content. This study employed a literature review to systematically examine the research status of key flaxseed dehulling technologies in China. It explored the characteristics, efficiencies, and quality differences among various dehulling methods, while also drawing on advanced techniques, such as chemical and ultrasonic dehulling, to provide new perspectives and theoretical support for flaxseed dehulling. Comprehensive analysis revealed that mechanical dehulling (the impact method and rolling and rubbing method) is the primary method used in China. The study also identified the issues in current flaxseed dehulling research in China and offers suggestions to guide future improvements and innovations in flaxseed processing, aiming to enhance the quality and nutritional value of flaxseed to meet diverse market demands. Full article

This research conducts an analysis of the sustainability of urban agriculture in Shanghai over the period 2010 to 2020, employing the Triple Bottom Line (TBL) concept as a framework to evaluate sustainability across economic, environmental, and social dimensions through the formulation and application of a comprehensive indicator system. Utilizing the Delphi method alongside the Analytic Hierarchy Process (AHP) for determining indicators and their respective weights, this study adopts a methodologically rigorous approach to analysis. The findings reveal an overall enhancement in agricultural sustainability, albeit accompanied by a decline in economic sustainability. Notably, environmental sustainability emerged as a paramount concern, underscoring the essentiality of incorporating environmental indicators within urban agricultural initiatives. The paper addresses significant challenges such as elevated land prices, demographic shifts, and the imperative for more stringent environmental regulations. It advocates for a multidimensional strategy integrating advanced agricultural technologies and cross-sectoral partnerships to bolster sustainability. Furthermore, the study accentuates the necessity of achieving equilibrium among economic feasibility, environmental stewardship, and social equity to pursue sustainable urban agriculture in Shanghai. Additionally, it highlights the critical role of strategic agricultural policy formulation in fostering sectoral resilience and ensuring enduring sustainability. Full article

Soil salinization has a significant impact on agricultural production and ecology. There is an urgent demand to establish an effective method that monitors the spatial and temporal distribution of soil salinity. In this study, a multi-indicator soil salinity monitoring model was proposed for monitoring soil salinity in Bachu County, Kashgar Region, Xinjiang, from 2002 to 2022. The model was established by combining multiple predictors (spectral, salinity, and composite indices and topographic factors) and the accuracy of the four models (Random Forest [RF], Partial Least Squares [PLS], Classification Regression Tree [CART], and Support Vector Machine [SVM]) was compared. The results reveal the high accuracy of the optimized prediction model, and the order of the accuracy is observed as RF > PLS > CART > SVM. The most accurate model, RF, exhibited an R² of 0.723, a root mean square error (RMSE) of 2.604 g·kg⁻¹, and a mean absolute error (MAE) of 1.95 g·kg⁻¹ at a 0–20 cm depth. At a 20–40 cm depth, RF had an R² value of 0.64, an RMSE of 3.62 g·kg⁻¹, and an MAE of 2.728 g·kg⁻¹. Spatial changes in soil salinity were observed throughout the study period, particularly increased salinization from 2002 to 2012 in the agricultural and mountainous areas within the central and western regions of the country. However, salinization declined from 2012 to 2022, with a decreasing trend in salinity observed in the top 0–20 cm of soil, followed by an increasing trend in salinity at a 20–40 cm depth. The proposed method can effectively extract large-scale soil salinity and provide a practical basis for simplifying the remote sensing monitoring and management of soil salinity. This study also provides constructive suggestions for the protection of agricultural areas and farmlands. Full article

Large amounts of spent mushroom substrate (SMS) are produced globally, but their utilization efficiency is low, which leads to negative environmental impacts, such as water, soil, and air pollution. SMS contains nutrients, such as cell proteins, with a potential application in animal feed. However, the lignocellulose in SMS restricts animal digestion and absorption, thus hindering its application in animal nutrition. We investigated the potential of cellulase, xylanase, β-galactosidase, and a variety of microorganisms to optimize the conditions for reducing sugars’ (RS) production and the degradation rate of neutral detergent fibers. The results showed that the optimum proportion of multiple enzymes for glucose production of up to 210.89 mg/g were 10% cellulase, 10% xylanase, and 2% β -galactosidase, at 50 °C and 60% moisture for a 20 h hydrolysis duration. To enhance the optimal enzymolysis combination, co-fermentation experiments with multiple microorganisms and enzymes showed that inoculation with 10% Bacillus subtilis, 2% Pediococcus acidilactici, and 2% Saccharomyces cerevisiae, in combination with 10% cellulase, 10% xylanase, 2% β-galactosidase, and 1% urea, at 36.8°C and 59% moisture for 70 h hydrolysis, could lead to a 23.69% degradation rate of the neutral detergent fiber. This process significantly increased the degradation rate of the neutral detergent fiber and the nutrient content of Pleurotus eryngii compared to the initial fermentation conditions. Overall, our study generated optimal co-fermentation conditions for bacteria and enzymes and provides a practical reference for biological feed synthesis using P. eryngii spent mushroom substrate. Full article

Straw return is one of the main methods for protecting black soil. Efficient and accurate straw return detection is important for the sustainability of conservation tillage. In this study, a rapid straw return detection method is proposed for large areas. An optimized Sh-DeepLabv3+ model based on the aforementioned detection method and the characteristics of straw return in Jilin Province was then used to classify plots into different straw return cover types. The model used Mobilenetv2 as the backbone network to reduce the number of model parameters, and the channel-wise feature pyramid module based on channel attention (CA-CFP) and a low-level feature fusion module (LLFF) were used to enhance the segmentation of the plot details. In addition, a composite loss function was used to solve the problem of class imbalance in the dataset. The results show that the extraction accuracy is optimal when a 2048 × 2048-pixel scale image is used as the model input. The total parameters of the improved model are 3.79 M, and the mean intersection over union (MIoU) is 96.22%, which is better than other comparative models. After conducting a calculation of the form–grade mapping relationship, the error value of the area prediction was found to be less than 8%. The results show that the proposed rapid straw return detection method based on Sh-DeepLabv3+ can provide greater support for straw return detection. Full article

Nowadays, machine learning (ML) is a useful technology due to its high accuracy in constructing non-linear models and algorithms that can adapt to the complexity and diversity of data. Thus, the current work aimed to predict the soil quality index (SQI) from extensive soil data, achieving high accuracy with the artificial neural networks (ANN) model. However, the efficiency of ANN depends on the accuracy of the data that is prepared for training. For this purpose, MATLAB programming language was used to enable the calculation, classification, and compilation of the results into databases within a few minutes. The proposed MATLAB program was highly efficient, accurate, and quick in calculating soil big data for training the machine compared with traditional methods. The database contains 306 vector sets, 80% of them are used for training and the remaining 20% are reserved for testing. The optimal model obtained comprises one hidden layer with 250 neurons and one output layer with a sigmoid function. The ANN achieved a high coefficient of determination (R²) values for SQI estimation, with around 0.97 and 0.98 for training and testing, respectively. The results indicate that 36.93% of the total soil samples belonged to the very high quality class (C1). In contrast, the high quality (C2), moderate quality (C3), low quality (C4), and very low quality (C5) classes accounted for 10.46%, 31.37%, 20.92%, and 0.33% of the samples, respectively. The high contents of CaCO₃, pH, sodium saturation, salinity, and clay content were identified as limiting factors in certain areas. The results of this study indicated high accuracy of soil quality assessment using physical, chemical, and fertility soil features in regression analysis with ANN. This method, which is suitable for arid zones, enhances agricultural productivity and decision-making by identifying critical soil quality categories and constraints. Full article

Performing joint genomic predictions for multiple breeds (MBGP) to expand the reference size is a promising strategy for improving the prediction for limited population sizes or phenotypic records for a single breed. This study proposes an MBGP model—mbBayesAB, which treats the same traits of different breeds as potentially genetically related but different, and divides chromosomes into independent blocks to fit heterogeneous genetic (co)variances. Best practices of random effect (co)variance matrix priors in mbBayesAB were analyzed, and the prediction accuracies of mbBayesAB were compared with within-breed (WBGP) and other commonly used MBGP models. The results showed that assigning an inverse Wishart prior to the random effect and obtaining information on the scale of the inverse Wishart prior from the phenotype enabled mbBayesAB to achieve the highest accuracy. When combining two cattle breeds (Limousin and Angus) in reference, mbBayesAB achieved higher accuracy than the WBGP model for two weight traits. For the marbling score trait in pigs, MBGP of the Yorkshire and Landrace breeds led to a 6.27% increase in accuracy for Yorkshire validation using mbBayesAB compared to that using the WBGP model. Therefore, considering heterogeneous genetic (co)variance in MBGP is advantageous. However, determining appropriate priors for (co)variance and hyperparameters is crucial for MBGP. Full article

The aim of this study was to determine the impact of different crop production systems (organic, integrated, and conventional) on the yields of several spring wheat (Triticum aestivum L.) cultivars. A field experiment was carried out at the Agricultural Experimental Station of the Institute of Soil Science and Plant Cultivation in Osiny (Poland) in three consecutive growing seasons (2014, 2015, and 2016). Two factors were included in the experiment: the crop production system (organic, integrated, and conventional) and spring wheat cultivars (Kandela, Izera, Ostka Smolicka, and Waluta). The crop production system significantly differentiated the yield, health, and weed infestation of the spring wheat. Wheat yield in the conventional system (6.12 t·ha⁻¹) was higher than in the organic system (3.68 t·ha⁻¹) by 67%, whereas, in the integrated system (7.61 t·ha⁻¹), it was greater than in the organic system by 109%. The lower yields in the organic system were mainly due to fewer ears per m² and a smaller 1000-grain weight. In the organic system, we also observed a higher infestation of wheat by foliar fungal pathogens and weeds compared with the conventional and integrated systems. The spring wheat cultivars differed in yield structure and resistance to infestation by fungal pathogens. The Waluta and Izera cultivars performed well in all systems but yielded the best in the integrated and conventional ones. The Kandela cultivar was the most suitable for the organic system, as it achieved the highest yield (4.16 t·ha⁻¹). This was mainly due to its ability to form a compact canopy with relatively high ear density, a large 1000-grain weight, and the highest resistance to fungal pathogens. The results for cultivars’ performance in the organic system may be useful for farmers in decreasing yield gaps in relation to integrated and conventional systems. Full article

To improve the speed and accuracy of the methods used for the recognition and positioning of strawberry plants, this paper is concerned with the detection of elevated-substrate strawberries and their picking points, using a strawberry picking robot, based on the You Only Look Once version 7 (YOLOv7) object detection algorithm and Red Green Blue-Depth (RGB-D) sensing. Modifications to the YOLOv7 model include the integration of more efficient modules, incorporation of attention mechanisms, elimination of superfluous feature layers, and the addition of layers dedicated to the detection of smaller targets. These modifications have culminated in a lightweight and improved YOLOv7 network model. The number of parameters is only 40.3% of that of the original model. The calculation amount is reduced by 41.8% and the model size by 59.2%. The recognition speed and accuracy are also both improved. The frame rate of model recognition is increased by 19.3%, the accuracy of model recognition reaches 98.8%, and [email protected] reaches 96.8%. In addition, we have developed a method for locating strawberry picking points based on strawberry geometry. The test results demonstrated that the average positioning success rate and average positioning time were 90.8% and 76 ms, respectively. The picking robot in the laboratory utilized the recognition and positioning method proposed in this paper. The error of hand–eye calibration is less than 5.5 mm on the X-axis, less than 1.6 mm on the Y-axis, and less than 2.7 mm on the Z-axis, which meets the requirements of picking accuracy. The success rate of the picking experiment was about 90.8%, and the average execution time for picking each strawberry was 7.5 s. In summary, the recognition and positioning method proposed in this paper provides a more effective method for automatically picking elevated-substrate strawberries. Full article

In order to solve the problem of the low pickup efficiency of the robotic arm when harvesting safflower filaments, we established a pickup trajectory cycle and an improved velocity profile model for the harvest of safflower filaments according to the growth characteristics of safflower. Bezier curves were utilized to optimize the picking trajectory, mitigating the abrupt changes produced by the delta mechanism during operation. Furthermore, to overcome the slow convergence speed and the tendency of the ant colony algorithm to fall into local optima, a safflower harvesting trajectory planning method based on an ant colony genetic algorithm is proposed. This method includes enhancements through an adaptive adjustment mechanism, pheromone limitation, and the integration of optimized parameters from genetic algorithms. An optimization model with working time as the objective function was established in the MATLAB environment, and simulation experiments were conducted to optimize the trajectory using the designed ant colony genetic algorithm. The simulation results show that, compared to the basic ant colony algorithm, the path length with the ant colony genetic algorithm is reduced by 1.33% to 7.85%, and its convergence stability significantly surpasses that of the basic ant colony algorithm. Field tests demonstrate that, while maintaining an S-curve velocity, the ant colony genetic algorithm reduces the harvesting time by 28.25% to 35.18% compared to random harvesting and by 6.34% to 6.81% compared to the basic ant colony algorithm, significantly enhancing the picking efficiency of the safflower-harvesting robotic arm. Full article

Emmental or Swiss cheese is a hard, ripened cheese appreciated by consumers for its appearance and taste. This study aimed to investigate the profile of Swiss cheese consumers and producers from Ţara Dornelor area, Romania, along with the sensory analysis of the Dorna Swiss cheese produced there. For this purpose, a questionnaire was applied to 268 participants to evaluate consumer behavior. Consumers were grouped depending on consumption frequency (low—once or a few times a year, medium—once a month, and high—once a week or more), and the behavior of groups was evaluated. Producer opinion was assessed by interview and Swiss cheese sensory characteristics in two seasons were determined by sensory analysis using a semi-trained panel. The results showed that the main factors affecting consumer purchase decision are the ingredients (4.43), taste and flavor (4.41), appearance and texture (4.23), producer (3.98), nutritional value (3.88), and product history (3.67). Clustering of consumers depending on consumption frequency revealed significant differences (p < 0.05) regarding the purchase place and some factors influencing the purchase decision such as price, health benefits, and nutritional value. Producers asserted that the quality of milk is the main problem in Swiss cheese production. They consider that the raw material quality and origin, hygiene, utilities, and legislation have the greatest impact on the production process, while the trading is mainly affected by the product taste and flavor, appearance and texture, quality label, price, and product history. The sensory characteristics differed significantly (p < 0.05) between producers and seasons, with the sample produced in a stainless-steel tank and without exogenous microflora being the most appreciated in summer. These results could help producers adapt their product quality and marketing policy to consumer preferences. Full article

This study aimed to review recent developments in the agri-food industry, focusing on the integration of innovative digital systems into the livestock industry. Over the last 50 years, the production of animal-based foods has increased significantly due to the rising demand for meat. As a result, farms have increased their livestock numbers to meet consumer demand, which has exacerbated challenges related to environmental sustainability, human health, and animal welfare. In response to these challenges, precision livestock farming (PLF) technologies have emerged as a promising solution for sustainable livestock production. PLF technologies offer farmers the opportunity to increase efficiency while mitigating environmental impact, securing livelihoods, and promoting animal health and welfare. However, the adoption of PLF technologies poses several challenges for farmers and raises animal welfare concerns. Additionally, the existing legal framework for the use of PLF technologies is discussed. In summary, further research is needed to advance the scientific understanding of PLF technologies, and stakeholders, including researchers, policymakers, and funders, need to prioritize ethical considerations related to their implementation. Full article

Meteorological conditions play a crucial role in driving outbreaks of rubber tree powdery mildew (RTPM). As the climate warms and techniques improve, rubber cultivation is expanding to higher latitudes, and the changing climate increases the RTPM risk. Rubber plantations on Hainan Island, situated on the northern margin of the tropics, have been selected as a case study to explore the meteorological mechanisms behind RTPM outbreaks quantitatively using a structural equation model, and project current and future RTPM outbreak patterns under different climate change scenarios by building predictive models based on data-driven algorithms. The following results were obtained: (1) days with an average temperature above 20 °C and days with light rain were identified as key meteorological drivers of RTPM using structural equation modeling (R² = 0.63); (2) the Bayesian-optimized least-squares boosted trees ensemble model accurately predicted the interannual variability in the historical RTPM disease index (R² = 0.79); (3) currently, due to the increased area of rubber plantations in the central region of Hainan, there is a higher risk of RTPM; and (4) under future climate scenarios, RTPM shows a decreasing trend (at a moderate level), with oscillating and sporadic outbreaks primarily observed in the central and northwest regions. We attribute this to the projected warming and drying trends that are unfavorable for RTPM. Our study is expected to enhance the understanding of the impact of climate change on RTPM, provide a prediction tool, and underscore the significance of the climate-aware production and management of rubber. Full article

As an important cereal crop, maize is a versatile and multi-purpose crop, primarily used as a feed globally, but also is important as a food crop, and has other uses such as oil and industrial raw materials. Quality detection is an indispensable part of functional and usage classification, avoiding significant waste as well as increasing the added value of the product. The research on algorithms for real-time, accurate, and non-destructive identification and localization of corn kernels based on quality classification and equipped with non-destructive algorithms suitable for embedding in intelligent agricultural machinery systems is a key step in improving the effective utilization rate of maize kernels. The difference in maize kernel quality leads to significant differences in price and economic benefits. This algorithm reduced unnecessary waste caused by the low efficiency and accuracy of manual and mechanical detection. Image datasets of four kinds of maize kernel quality were established and each image contains a total of about 20 kernels of different quality randomly distributed. Based on the self-built dataset, the YOLOv7-tiny, as the backbone network, was used to design a maize kernel detection and recognition model named “YOLOv7-MEF”. Firstly, the backbone feature layer of the algorithm was replaced by MobileNetV3 as the feature extraction backbone network. Secondly, ESE-Net was used to enhance feature extraction and obtain better generalization performance. Finally, the loss function was optimized and replaced with the Focal-EOIU loss function. The experiment showed that the improved algorithm achieved an accuracy of 98.94%, a recall of 96.42%, and a Frame Per Second (FPS) of 76.92 with a model size of 9.1 M. This algorithm greatly reduced the size of the model while ensuring high detection accuracy and has good real-time performance. It was suitable for deploying embedded track detection systems in agricultural machinery equipment, providing a powerful theoretical research method for efficient detection of corn kernel quality. Full article

Sustainable Development Goal (SDG) 2 seeks to end hunger and guarantee food and nutrition security worldwide by 2030. Smallholder irrigation development remains a key strategy to achieve SDG 2. This study assesses how smallholder irrigation contributes to household food security in Mberengwa district, Zimbabwe. Primary data were gathered from a randomly chosen sample of 444 farmers (344 irrigators and 100 non-irrigators) using a structured questionnaire. Microsoft Excel and Statistical Package for Social Sciences version 27 software packages were used to analyse the data. Descriptive statistics, chi-square test, t-test, and binary logistic regression were performed. The t-test results show significant differences in mean between irrigators and non-irrigators for household size, the dependency ratio, farming experience, farm income, food expenditure share, and livestock owned (p < 0.05). Irrigators had significantly higher area planted, yield, and quantity sold for maize during the summer than non-irrigators (p < 0.05). Food Consumption Score results show that 97% of irrigators and 45% of non-irrigators were food secure. Binary logistic regression results reveal a significant association between food security and household size, irrigation access, and farm income (p < 0.05). In conclusion, access to smallholder irrigation increases household food security. The government and its development partners should prioritise investments in smallholder irrigation development, expansion, and rehabilitation. Full article