Topics
| Filter results17 paper(s) found. |
|---|
1. Geographic Database in Precision Agriculture for the Development of AI ResearchAgriculture 4.0 has profoundly transformed production processes by incorporating technologies such as Precision Agriculture, Artificial Intelligence, the Internet of Things, and telemetry. This evolution has enabled more accurate and timely decision-making in agriculture. In response to this movement, the Precision Agriculture Laboratory (AgriLab) of UTFPR, located in Medianeira, proposes the establishment of a consistent and standardized database. This database is continually updated with su... E.N. Avila, C.L. Bazzi, W.K. Oliveira, K. Schenatto, R. Sobjak, D.M. Rocha |
2. Explainable Neural Network Alternatives for Ai Predictions: Genetic Algorithm Quantitative Association Rule MiningNeural networks in one form or another are common precision agriculture artificial intelligence techniques for making predictions based on data. However, neural networks are computationally intensive to train and to run, and are typically “black-box” models without explainable output. This paper investigates an alternative artificial intelligence prediction technique, genetic algorithm quantitative association rule mining, which creates explainable output with impacts directly qua... M. Everett |
3. Sampling Bumble Bees and Floral Resources Using Deep Learning and UAV ImageryPollinators, essential components of natural and agricultural systems, forage over relatively large spatial scales. This is especially true of large generalist species, like bumble bees. Thus, it can be difficult to estimate the amount and diversity of floral resources available to them. Floral cover and diversity are often estimated over large areas by extrapolation from small scale samples (e.g., a 1-m quadrat) but the accuracy of such estimates can vary depending on the spatial patchiness ... B. Spiesman, I. Grijalva, D. Holthaus, B. Mccornack |
4. Optimizing Nitrogen Application in Global Wheat Production by an Integrated Bayesian and Machine Learning ApproachWheat production plays a pivotal role in global food security, with nitrogen fertilizer application serving as a critical factor. The precise application of nitrogen fertilizer is imperative to maximize wheat yield while avoiding environmental degradation and economic losses resulting from excess or inadequate usage. The integration of Bayesian and machine learning methodologies has gained prominence in the realm of agricultural research. Bayesian and machine learning based methods have great... Z. Liu, X. Liu, Y. Tian, Y. Zhu, W. Cao, Q. Cao |
5. Automated Southern Leaf Blight Severity Grading of Corn Leaves in RGB Field ImageryPlant stress phenotyping research has progressively addressed approaches for stress quantification. Deep learning techniques provide a means to develop objective and automated methods for identifying abiotic and biotic stress experienced in an uncontrolled environment by plants comparable to the traditional visual assessment conducted by an expert rater. This work demonstrates a computational pipeline capable of estimating the disease severity caused by southern corn leaf blight in images of ... C. Ottley, M. Kudenov, P. Balint-kurti, R. Dean, C. Williams |
6. Deep Learning for Predicting Yield Temporal Stability from Short Crop RotationsInvestigating the temporal stability of yield in management zones is crucial for both producers and researchers, as it helps in mitigating the adverse impacts of unpredictable disruptions and weather events. The diversification of cropping systems is an approach which leads to reduced variability in yield while improving overall field resilience. In this six-year study spanning from 2016 to 2021, we monitored 40 distinct fields owned by 10 producers situated in Quebec, Canada. These... E. Lord, A.A. Boatswain jacques, A.B. Diallo, M. Khakbazan, A. Cambouris |
7. Enhancing Agricultural Feedback Analysis Through VUI and Deep Learning IntegrationA substantial amount of information relies on consumers, influencing aspects from product adoption to overall satisfaction. Similarly, the agricultural sector is entirely dependent on farmers, who dictate the success of products and highlight associated challenges. Our study aligns with this perspective, recognizing the significance of understanding farmers' needs to assist tractor manufacturing industries. As these industries aim for widespread adoption of their products among farmers, i... S. Kaushal, A. Sharda |
8. An Open Database of Crop Yield Response to Fertilizer Application for SenegalFood security is one of the major global challenges today. Africa is one of the continents with the largest gaps in terms of challenges for food security. In Senegal, about 60% of the population resides in rural areas and the cropping systems are characterized as a low productivity system, low input and in reduced areas, smallholder subsistence systems. Increasing crop productivity would have a positive impact on food security in this country. One of the main factors limiting crop produ... F. Gomez, A. Carcedo, A. Diatta, L. Nagarajan, V. Prasad, Z. Stewart, S. Zingore, I. Ciampitti, P. Djighaly |
9. On Data-driven Crop Yield Modelling, Predicting, and Forecasting and the Common Flaws in Published StudiesThere has been a recent surge in the number of studies that aim to model crop yield using data-driven approaches. This has largely come about due to the increasing amounts of remote sensing (e.g. satellite imagery) and precision agriculture data available (e.g. high-resolution crop yield monitor data), and abundance of machine learning modelling approaches. This is a particular problem in the field of Precision Agriculture, where many studies will take a crop yield map (or a small number), cr... P. Filippi, T. Bishop, S. Han, I. Rund |
10. Generative Modeling Method Comparison for Class Imbalance CorrectionAn image dataset, for use in object detection of hay bales, with over 6000 images of both good and bad hay bales was collected. Unfortunately, the dataset developed a class imbalance, with more good bale images than bad bales. This dataset class imbalance caused the bad bale class to over train and the good bale class to under train, severely impacting precision, and recall. To correct this imbalance and provide a comparison of differing generative modeling methods; three di... B. Vail, Z. Oster, B. Weinhold |
11. Deep Learning to Estimate Sorghum Yield with Uncrewed Aerial System ImageryIn the face of growing demand for food, feed, and fuel, plant breeders are challenged to accelerate yield potential through quick and efficient cultivar development. Plant breeders often conduct large-scale trials in multiple locations and years to address these goals. Sorghum breeding, integral to these efforts, requires early, accurate, and scalable harvestable yield predictions, traditionally possible only after harvest, which is time-consuming and laborious. This research harnesses high-t... M.A. Bari, A. Bakshi, T. Witt, D. Caragea, K. Jagadish, T. Felderhoff |
12. Machine Vision in Hay Bale ProductionThe goal of this project is to develop a system capable of real-time detection, pass/fail classification, and location tracking of large square hay bales under field conditions. First, a review of past and current methods of object detection was carried out. This led to the selection of the YOLO family of detectors for this project. The image dataset was collected through help from our sponsor, collection of images from the K-STATE research farm, and images collected from th... B. Vail |
13. Design of an Autonomous Ag Platform Capable of Field Scale Data Collection in Support of Artificial IntelligenceThe Pivot+ Array is intended to serve as an innovative, multi-user research platform dedicated to the autonomous monitoring, analysis, and manipulation of crops and inputs at the plant scale, covering extensive areas. It will effectively address many constraints that have historically limited large-scale agricultural sensor and robotic research. This achievement will be made possible by augmenting the well-established center pivot technology, known for its autonomy, with robust power inf... S. Jha, J. Krogmeier, D. Buckmaster, D.J. Love, R.H. Grant, M. Crawford, C. Brinton, C. Wang, D. Cappelleri, A. Balmos |
14. Wheat Spikes Counting Using Density Prediction Convolution Neural NetworkVision-based wheat spikes counting can be valuable for pre-harvest yield estimation for growers and researchers. In this study, wheat spike counting convolutions neural networks were implemented to solve the problem of vision-based wheat yield prediction problem. Encoder-decoder style convolutional neural networks (CNN) were developed with a Global Sum Pooling (GSP) layer as its output layer and trained to produce a density map which predicts the pixelwise wheat spikes density. Thi... C. Liew, S. Pitla |
15. Simultaneously Estimating Crop Biomass and Nutrient Parameters Using UAS Remote Sensing and Multitask LearningRapid and accurate estimation of crop growth status and nutrient levels such as aboveground biomass, nitrogen, phosphorus, and potassium concentrations and uptake is critical with respect to precision agriculture and field-based crop monitoring. Recent developments in Uncrewed Aircraft Systems (UAS) and sensor technologies have enabled the collection of high spatial, spectral, and temporal remote sensing data over large areas at a lower cost. Coupled deep learning-based modeling approaches wi... P. Kovacs, M. Maimaitijiang, B. Millett, L. Dorissant, I. Acharya, U.U. Janjua, K. Dilmurat |
16. Potato Disease Detection Using Laser Speckle Imaging and Deep LearningEarly detection of potato diseases is essential for minimizing crop loss. Implementing advanced imaging techniques can significantly improve the accuracy and efficiency of disease detection in potato crops. Leveraging machine learning algorithms can further enhance the speed and precision of disease identification, enabling timely intervention measures. This work presents a novel potato disease detection technique using whole-potato speckle imaging and deep learning. Laser Speckle Imaging (LS... A.H. Rabia, M.A. Salem |
17. Application of Advanced Soft Computing to Estimate Potato Tuber Yield: a Case Study from Atlantic CanadaThe potato crop plays a crucial role in the economy of Atlantic Canada, particularly in Prince Edward Island and New Brunswick, where it contributes significantly to potato production. To help farmers make informed decisions for sustainable and profitable farming, this study was conducted to examine the variations in potato tuber yield based on thirty soil properties collected over four growing seasons through experimental trials. The study employed an advanced and explainable ensemble model ... Q.U. Zaman, A. Farooque, M. Jamei, T.J. Esau |