Nyéki Anikó, Kerepesi Csaba, Daróczy Bálint, Benczúr András, Milics Gábor, Nagy J., Harsányi Endre, Kovács Attila József, Neményi Miklós
Application of spatio‑temporal data in site‑specific maize yield prediction with machine learning methods
Precision Agriculture (2021) 22:1397–1415
https://doi.org/10.1007/s11119-021-09833-8
Abstract
In order to meet the requirements of sustainability and to determine yield drivers and limiting factors, it is now more likely that traditional yield modelling will be carried out using artifcial intelligence (AI). The aim of this study was to predict maize yields using AI that uses spatio-temporal training data. The paper has advanced a new method of maize yield prediction, which is based on spatio-temporal data mining. To fnd the best solution, various models were used: counter-propagation artifcial neural networks (CP-ANNs), XYfused Querynetworks (XY-Fs), supervised Kohonen networks (SKNs), neural networks with Rectangular Linear Activations (ReLU), extreme gradient boosting (XGBoost), support-vector machine (SVM), and diferent subsets of the independent variables in fve vegetation periods. Input variables for modelling included: soil parameters (pH, P2O5, K2O, Zn, clay content, ECa, draught force, Cone index), micro-relief averages, and meteorological parameters for the 63 treatment units in a 15.3 ha research feld. The best performing method (XGBoost) reached 92.1% and 95.3% accuracy on the training and the test sets. Additionally, a novel method was introduced to treat individual units in a lattice system. The lattice-based smoothing performed an additional increase in Area under the curve (AUC) to 97.5% over the individual predictions of the XGBoost model. The models were developed using 48 diferent subsets of variables to determine which variables consistently contributed to prediction accuracy. By comparing the resulting models, it was shown that the best regression model was Extreme Gradient Boosting Trees, with 92.1% accuracy (on the training set). In addition, the method calculates the infuence of the spatial distribution of site-specifc soil fertility on maize grain yields. This paper provides a new method of spatio-temporal data analyses, taking the most important infuencing factors on maize yields into account.
Keywords Maize yield prediction · Machine learning methods · Gradient boosting (XGBoost) · Random felds · Yield infuencing variables · Decision support in crop production
László Moldvai, Péter Ákos Mesterházi, Gergely Teschner, Anikó Nyéki
Weed Detection and Classification with Computer Vision Using a Limited Image Dataset
Applied Sciences 2024, 14 ,4839.
https://doi.org/10.3390/app14114839
In agriculture, as precision farming increasingly employs robots to monitor crops, the use of weeding and harvesting robots is expanding the need for computer vision. Currently, most researchers and companies address these computer vision tasks with CNN-based deep learning. This technology requires large datasets of plant and weed images labeled by experts, as well as substantial computational resources. However, traditional feature-based approaches to computer vision can extract meaningful parameters and achieve comparably good classification results with only a tenth of the dataset size. This study presents these methods and seeks to determine the minimum number of training images required to achieve reliable classification. We tested the classification results with 5, 10, 20, 40, 80, and 160 images per weed type in a four-class classification system. We extracted shape features, distance transformation features, color histograms, and texture features. Each type of feature was tested individually and in various combinations to determine the best results. Using six types of classifiers, we achieved a 94.56% recall rate with 160 images per weed. Better results were obtained with more training images and a greater variety of features.
AMBRUS BÁLINT
APPLICATION POSSIBILITIES OF ROBOT TECHNIQUE IN ARABLE PLANT PROTECTION.
Széchenyi István University, Albert Kázmér Faculty of Agricultural and Food Sciences of Széchenyi István University, Department of Biosystems and Precision Technologies, Mosonmagyaróvár
Acta Agronomica Óváriensis Vol. 62. No.1. (2021)
SUMMARY
The author presents the characteristics of robots used in agriculture, including plant protection, as well as the possibilities of robot technology and the possibilities provided by robot technology. It places particular emphasis on the analysis of sustainability criteria. The most critical factor is the reduction of the use of synthetic pesticides and chemicals replacement by physical procedures. This knowledge helps the paradigm shift in the automatization and robotization of agricultural work. The future belongs to small, smart, interconnected, light machines. However, their properties raise moral and, within that, a number of ethical, management and social issues, both in the operation and in the design and manufacture of the machines working in the swarm. The future also belongs to the super-intelligent machines in this field, which will be able to improve themselves including software and hardware too. The dissertation also draws attention to the fact that a paradigm shift is needed in the sense that basic machines must be standardized, ie the design of a robot should not start from the basics, but can also use previously developed applications. A good example of this is From Toy to Tool: FTtT. Industrial robots offer many opportunities for innovation, but at the same time adapting this knowledge and experience to the natural environment poses serious challenges. Due to the specific nature robotics in the industry can only be partially utilized here, and in most of the individual production areas special solutions are also needed. It is likely that the robotization of agriculture will also require the development of new organizational and organizational structures. The article also helps with this.
Keywords: robot, robotics, plant protection, GPS, RTK, MI, drone
Tarek Alahmad, Miklós Neményi, Anikó Nyéki
“Applying IoT Sensors and Big Data to Improve Precision Crop Production: A Review”
Agronomy 2023, 13, 2603.
https://doi.org/10.3390/agronomy13102603
Széchenyi István University, Albert Kázmér Faculty of Agricultural and Food Sciences of Széchenyi István University, Department of Biosystems and Precision Technologies, Mosonmagyaróvár 9200, Hungary.
The potential benefits of applying information and communication technology (ICT) in precision agriculture to enhance sustainable agricultural growth were discussed in this review article. The current technologies, such as the Internet of Things (IoT) and artificial intelligence (AI), as well as their applications, must be integrated into the agricultural sector to ensure long-term agricultural productivity. These technologies have the potential to improve global food security by reducing crop output gaps, decreasing food waste, and minimizing resource use inefficiencies. The importance of collecting and analyzing big data from multiple sources, particularly in situ and on-the-go sensors, is also highlighted as an important component of achieving predictive decision making capabilities in precision agriculture and forecasting yields using advanced yield prediction models developed through machine learning. Finally, we cover the replacement of wired-based, complicated systems in infield monitoring with wireless sensor networks (WSN), particularly in the agricultural sector, and emphasize the necessity of knowing the radio frequency (RF) contributing aspects that influence signal intensity, interference, system model, bandwidth, and transmission range when creating a successful Agricultural Internet of Thing Ag-IoT system. The relevance of communication protocols and interfaces for presenting agricultural data acquired from sensors in various formats is also emphasized in the paper, as is the function of 4G, 3G, and 5G technologies in IoT-based smart farming. Overall, these research sheds light on the significance of wireless sensor networks and big data in the future of precision crop production.
Bálint Ambrus, Miklós Neményi, Attila Kovács, Anikó Nyéki
Development of a Small-Smart Robot in Mosonmagyaróvár
MEZŐGAZDASÁGI TECHNIKA
In recent times, several studies have highlighted the need for another paradigm shift, as the current negative impacts of agriculture on the biosphere cannot be mitigated using knowledge from traditional experiment-based research. Global food security is threatened by several factors, such as population growth, meat consumption trends, and the effects of climate change. Additionally, increasing pest, disease, and weed tolerance are placing greater pressure on both traditional and precision technologies. To alleviate the burden of these challenges, there is an opportunity to automate and robotize certain aspects of the farming process.
Zoltán Molnár, Mutum Lamnganbi, Wogene Solomon, Tibor Janda
Chitosan and cyanobacterial biomass accounting physiological and biochemical development of winter wheat (Triticum aestivum L.) under nutrient stress conditions
Agrosystems, Geosciences & Environment, 6, e20428 (2023)
https://doi.org/10.1002/agg2.20428
In the spirit of returning to nature and using science to increase crop productivity without posing any threat to the environment, researchers are paying attention to making natural products alternative sources of nutrients for plants at affordable prices. On top of this, chitosan and cyanobacteria have become popular in agriculture as metabolic enhancers, biofertilizers, and antimicrobial properties. Cyanobacteria are known to possess bio stimulating properties, whereas chitosan is well known for its inherent biological properties. To minimize nitrogen application, this experiment was conducted for the first time to check whether application of chitosan, microalgae or both with 50% nitrogen can balance the nutrient requirement for different physiological and biochemical development as effectively as a 100% nitrogen dose. The data were recorded only for the early vegetative stages, as the seeds were non-vernalized. The basic parameters recorded were hexose content, chlorophyll a, chlorophyll b, total phenol content (TPC) and relative water content (RWC). In most parameters, comparable results were found between the control (with a 100% recommended nitrogen dose) and other treatments (where either microalga, chitosan, or both were added). Whereas it was clearly shown that 50% of recommended nitrogen doses reduce the hexose, chlorophyll, and relative water contents. Thus, the treatments were effective in supplementing the developmental requirements. Therefore, the combined use of chitosan and cyanobacteria in crops significantly reduces nitrogen fertilization, increases photosynthesis, increases resistance to water stress, and increases antioxidant activity in modern agriculture.
Levente Vörös, Rita Ledóné Ábrahám
Effect of azadirachtin applied as seed dressing on the larval density of and root injury caused by the western corn rootworm/Diabrotica virgifera virgifera
J Plant Dis Prot 130, 757–767 (2023)
https://doi.org/10.1007/s41348-023-00763-3
The western corn rootworm (Diabrotica virgifera virgifera LeConte) is one of the most important pests of maize in Hungary. As both larvae and imagoes are capable of causing major economic losses, their control in continuous maize cropping systems is essential. The control of larvae is costly and the related use of large doses of soil disinfectants places an increased burden on the environment. In recent years, several chemical products used as soil insecticides and seed dressings have been phased out, thus increasing the value of environmentally friendly biological products that provide effective protection against the pest. The active ingredient azadirachtin, the extract of the seeds of Azadirachta indica is one of such biological agents. In our experiments, we studied the efficacy of two azadirachtin products, Neemazal T/S (1% azadirachtin; 10 g/l) and Neemazal F (5% azadirachtin; 50 g/l) used as seed dressing against western corn rootworm larvae. The products were used in different concentrations (10–150%) in different regions and on various soil-types in Hungary. The active ingredient could effectively control the pest in its larval stage. Treatments with concentrations exceeding 50% were effective in all the replications.
Antonella Dalle Zotte , Yazavinder Singh, Eszter Zsedely, Barbara Contiero, Bianca Palumbo, Marco Cullere
Dietary inclusion of defatted silkworm (Bombyx mori L.) pupa meal in broiler chickens: phase feeding effects on nutritional and sensory meat quality
Poultry Science 103, 7, 103812
https://www.sciencedirect.com/science/article/pii/S0032579124003912?via%3Dihub
The present experiment was conducted to test the effect of a 4% defatted silkworm (Bombyx mori) pupae meal (SWM) incorporation into chickens’ diets at different growth phases on meat quality characteristics and sensory traits. Ninety ROSS 308 day-old male broiler chickens were randomly assigned to 3 dietary groups, with 5 replicated pens/diet: the first group received a control (C) diet throughout the growing period of 42 d, the second group received a diet with 4% SWM (SWM1) during the starter phase (1–10 d) and the C diet up to slaughter, whereas the third group was fed the C diet during the starter phase and 4% SWM during the grower and finisher phases (SWM2). Diets were isonitrogenous and isoenergy, and birds had free access to feed and water throughout the experimental trial. At 42 d of age, 15 chickens/treatment were slaughtered at a commercial abattoir. Fatty acid (FA) and amino acid (AA) profiles and contents of meat, as well as its oxidative status, were determined in both breast and leg meat cuts. Also, a descriptive sensory analysis was performed on breast meat by trained panelists. Results highlighted that the SWM2 treatment increased the n-3 proportion and content in both breast and leg meat, thereby improving the omega-6/omega-3 (n-6/n-3) ratio in both cuts (P < 0.001). However, the dietary treatment had no significant effect on the oxidative status of either breast or leg meat (P > 0.05). The SWM had a limited impact on overall sensory traits of breast meat, but it contributed to improve meat tenderness in SWM-fed chickens (P < 0.01). Furthermore, SWM1 meat exhibited higher juiciness (P < 0.05) and off flavor intensity (P < 0.05) compared to the control meat. Overall, the present experiment indicated that defatted SWM holds promise as an alternative ingredient in chicken rations, ensuring satisfactory meat quality. Furthermore, administering SWM during the grower-finisher phase demonstrated beneficial effects on meat healthiness, ultimately enhancing n-3 fatty acids content and reducing the n-6/n-3 ratio.
János Tenke, Orsolya Vida, István Nagy, János Tossenberger
Classifying Genetic Lines in Pork Production by Ileal Crude Protein and Amino Acid Digestibility in Growing Pigs Animals
(Basel) 2023 Jun 6;13(12):1898.
https://doi.org/10.3390/ani13121898
https://www.mdpi.com/2076-2615/13/12/1898
The first aim of the study was to evaluate the effect of different dietary lysine (LYS) to energy (DE) ratios on the apparent ileal digestibility (AID) of crude protein (CP) and selected amino acids (AA) in growing pigs (40-60 kg) of different genotypes. The second aim was to classify genotypes into groups based on the AID of CP and AAs. The trials were conducted on a total of 90 cross-bred barrows (30 animals/genotype) in two replicates. Before the trial series, the experimental animals (average initial body weight (BW) = 40.9 ± 8.5 kg) were surgically fitted with post valve T-cannula (PVTC). The diets were formulated with six different total LYS/DE ratios. Titanium dioxide (TiO2) was added to the diets (5 g/kg) as an indigestible marker. Based on our results, it can be concluded that the LYS/DE ratio of the diets affected the AID of the CP and AA in different ways by each genotype (p < 0.05). It can also be concluded that pigs of different genetic potential can be classified with a high accuracy (91.7%) in respect of their CP and AA digestive capacity. Our results indicate the development of genetic-profile-based swine nutrition technologies as a future direction.
Induced and field mechanical effects on the hatchability of broiler breeder hatching eggs
Induzierte und feldmechanische Effekte auf den Bruterfolg von Bruteiern von Broiler-Elterntieren
Europ.Poult.Sci., 88. 2024, ISSN 1612-9199, © Verlag Eugen Ulmer, Stuttgart. DOI: 10.1399/eps.2024.397
Egg transport and rough egg handling can have numerous negative effects on hatchability. The authors monitored mechanical effects under field conditions by acceleration sensors and then simulated the same scale effects using a modelling machine to verify the effects on hatchability. To measure and record mechanical effects-instantaneous acceleration (m/s 2), three-dimensional HOBO® Pendant® G Data Logger were used.
The RSS, RSM values calculated from the data from the accelerometer can serve as a point of reference for practitioners to see the mechanical effect level which results in significant negative impact on the hatchability results. It was also revealed that the measurement of the recorded values in the different directions and the minimum and maximum values are important too. Using the HOBO® Pendant® G Data Logger and detailed logging (the exact location of the logger at the time of the technological steps at a given time) can reveal the location of the maximum impact. By analysing RMS x, y, z, the type of impact can be determined. By combining these two pieces of two information, the technological failure can be clearly revealed and corrected. The measurement process described by the authors provides practical advice for hatching egg producers. Moreover, attention is drawn to the short-term damage effect on hatchability, since the 5-minute treatment at 20 Hz, prior incubation significantly reduced the hatchability (P < 0.05), which was achieved at the level of 10.02 RSS m/s 2 and 12.3 m/s 2 maximum value in the direction of x-axis.
It is important for hatching egg producers to be aware that the damage to the mechanical effect is not only visible (broken, cracked eggshell), but can also negatively affect hatchability and thus the profitability of the sector. Furthermore, the typical “spider web” crack on the eggshell clearly refers to the mechanical impact caused by vibration.
Judit Márton, Ferenc Szabó
Some Actualities and Challenges in Sustainable Beef Cattle Breeding and Husbandry
Chemical Engineering Ttransactions, 107, 2023, 241-246., DOI:10.3303/CET2310704
Beef cattle farming is an environmentally friendly food-producing animal husbandry sector that is based largely on pasture and arable by-product feedstuffs. It faces several problems that need to be addressed from a sustainability point of view. Population growth and growing food demand raise concerns about the environmental consequences of expanding beef production using current systems. Consequently, this condition underscores the importance of maintaining an equilibrium between these sustainability pillars and the necessity of adopting more sustainable models. This review article analyzes the most important, recent literary sources dealing with the sustainability of the sector. As a result of the literary synthesis, it has been established that most experts emphasize the two main pillars of sustainability, namely, economic and environmental aspects. The present work directs attention to the third and possible fourth point, the social as well as the cultural aspects of the sustainability of the beef cattle sector, which will be increasingly important to keep in mind in the future.
Tamás Csürhés, Ferenc Szabó, Gabriella Holló, Edit Mikó, Márton Török,Szabolcs Bene
Relationship between Some Myostatin Variants and Meat Production Related Calving, Weaning and Muscularity Traits in Charolais Cattle
Animals 2023, 13, 1895. https://doi.org/10.3390/ani13121895
The slaughter value on live cattle can be assessed during visual conformation scoring, as well as by examining different molecular genetic information, e.g. myostatin gene, which can be responsible for muscle development. In this study F94L, Q204X, nt267, nt324 and nt414 alleles of the myostatin gene (MSTN) were examined in relation to birth weight (BIW), calving ease (CAE), 205-day weaning weight (CWW), muscle score of shoulder (MSS), muscle score of back (MSB), muscle score of thigh (MST), roundness score of thigh (RST), loin thickness score (LTS), and overall muscle development percentage (OMP) of Charolais weaned calves in Hungary. Multi-trait analysis of variance (GLM) and weighted linear regression analysis were used to process the data. Calves carrying the Q204X allele in heterozygous form achieved approximately 0.14 points higher MSB, MST and LTS and 1.2% higher OMP and gained 8.56 kg more CWW than their counterparts not carrying the allele (p<0.05). As for the F94L allele, there was a difference of 4.08 kg in CWW of the heterozygous animals, but this difference could not be proved statistically. The other alleles had no significant effect on the evaluated traits.