Models for predicting coffee yield from chemical characteristics of soil and leaves using machine learning

BACKGROUND

Coffee farming constitutes a substantial economic resource, representing a source of income for several countries due to the high consumption of coffee worldwide. Precise management of coffee crops involves collecting crop attributes (characteristics of the soil and the plant), mapping, and applying inputs according to the plants' needs. This differentiated management is precision coffee growing and it stands out for its increased yield and sustainability.

RESULTS

This research aimed to predict yield in coffee plantations by applying machine learning methodologies to soil and plant attributes. The data were obtained in a field of 54.6 ha during two consecutive seasons, applying varied fertilization rates in accordance with the recommendations of soil attribute maps. Leaf analysis maps also were monitored with the aim of establishing a correlation between input parameters and yield prediction. The machine-learning models obtained from these data predicted coffee yield efficiently. The best model demonstrated predictive fit results with a Pearson correlation of 0.86. Soil chemical attributes did not interfere with the prediction models, indicating that this analysis can be dispensed with when applying these models.

CONCLUSION

These findings have important implications for optimizing coffee management and cultivation, providing valuable insights for producers and researchers interested in maximizing yield using precision agriculture. © 2024 Society of Chemical Industry.

C3PO: a crop planning and production process ontology and knowledge graph

Vegetable crop farmers diversify their production by growing a range of crops during the season on the same plot. Crop diversification and rotation enables farmers to increase their income and crop yields while enhancing their farm sustainability against climatic events and pest attacks. Farmers must plan their agricultural work per year and over successive years. Planning decisions are made on the basis of their experience regarding previous plans. For the purpose of assisting farmers in planning decisions and monitoring, we developed the Crop Planning and Production Process Ontology (C3PO), i.e., a representation of agricultural knowledge and data for diversified crop production. C3PO is composed of eight modules to capture all crop production dimensions and complexity for representing farming practices and constraints. It encodes agricultural processes and farm plot organization and captures common agricultural knowledge. C3PO introduces a representation of technical itineraries, i.e., sequences of technical farming tasks to grow vegetables, from soil identification and seed selection to harvest and storage. C3PO is the backbone of a knowledge graph which aggregates data from heterogeneous related semantic resources, e.g., organism taxonomies, chemicals, reference crop listings, or development stages. C3PO and its knowledge graph are used by the Elzeard enterprise to develop knowledge-based decision support systems for farmers. This article describes how we built C3PO and its knowledge graph-which are both publicly available-and briefly outlines their applications.

Ecosystem Services at the Farm Level-Overview, Synergies, Trade-Offs, and Stakeholder Analysis

The current geological epoch is characterized by anthropogenic activity that greatly impacts on natural ecosystems and their integrity. The complex networks of ecosystem services (ESs) are often ignored because the provision of natural resources, such as food and industrial crops, is mistakenly viewed as an independent process separate from ecosystems and ignoring the impacts on ecosystems. Recently, research has intensified on how to evaluate and manage ES to minimize environmental impacts, but it remains unclear how to balance anthropogenic activity and ecosystem integrity. This paper reviews the main ESs at farm level including provisioning, regulating, habitat, and cultural services. For these ESs, synergies are outlined and evaluated along with the respective practices (e.g., cover- and intercropping) and ES suppliers (e.g., pollinators and biocontrol agents). Further, several farm-level ES trade-offs are discussed along with a proposal for their evaluation. Finally, a framework for stakeholder approaches specific to farm-level ES is put forward, along with an outlook on how existing precision agriculture technologies can be adapted for improved assessment of ES bundles. This is believed to provide a useful framework for both decision makers and stakeholders to facilitate the development of more sustainable and resilient farming systems.

Toward the Next Generation of Digitalization in Agriculture Based on Digital Twin Paradigm

Digitalization has impacted agricultural and food production systems, and makes application of technologies and advanced data processing techniques in agricultural field possible. Digital farming aims to use available information from agricultural assets to solve several existing challenges for addressing food security, climate protection, and resource management. However, the agricultural sector is complex, dynamic, and requires sophisticated management systems. The digital approaches are expected to provide more optimization and further decision-making supports. Digital twin in agriculture is a virtual representation of a farm with great potential for enhancing productivity and efficiency while declining energy usage and losses. This review describes the state-of-the-art of digital twin concepts along with different digital technologies and techniques in agricultural contexts. It presents a general framework of digital twins in soil, irrigation, robotics, farm machineries, and food post-harvest processing in agricultural field. Data recording, modeling including artificial intelligence, big data, simulation, analysis, prediction, and communication aspects (e.g., Internet of Things, wireless technologies) of digital twin in agriculture are discussed. Digital twin systems can support farmers as a next generation of digitalization paradigm by continuous and real-time monitoring of physical world (farm) and updating the state of virtual world.

Is Seeing Still Believing? Leveraging Deepfake Technology for Livestock Farming

Deepfake technologies are known for the creation of forged celebrity pornography, face and voice swaps, and other fake media content. Despite the negative connotations the technology bears, the underlying machine learning algorithms have a huge potential that could be applied to not just digital media, but also to medicine, biology, affective science, and agriculture, just to name a few. Due to the ability to generate big datasets based on real data distributions, deepfake could also be used to positively impact non-human animals such as livestock. Generated data using Generative Adversarial Networks, one of the algorithms that deepfake is based on, could be used to train models to accurately identify and monitor animal health and emotions. Through data augmentation, using digital twins, and maybe even displaying digital conspecifics (digital avatars or metaverse) where social interactions are enhanced, deepfake technologies have the potential to increase animal health, emotionality, sociality, animal-human and animal-computer interactions and thereby productivity, and sustainability of the farming industry. The interactive 3D avatars and the digital twins of farm animals enabled by deepfake technology offers a timely and essential way in the digital transformation toward exploring the subtle nuances of animal behavior and cognition in enhancing farm animal welfare. Without offering conclusive remarks, the presented mini review is exploratory in nature due to the nascent stages of the deepfake technology.

Sensing Architecture for Terrestrial Crop Monitoring: Harvesting Data as an Asset

Very often, the root of problems found to produce food sustainably, as well as the origin of many environmental issues, derive from making decisions with unreliable or inexistent data. Data-driven agriculture has emerged as a way to palliate the lack of meaningful information when taking critical steps in the field. However, many decisive parameters still require manual measurements and proximity to the target, which results in the typical undersampling that impedes statistical significance and the application of AI techniques that rely on massive data. To invert this trend, and simultaneously combine crop proximity with massive sampling, a sensing architecture for automating crop scouting from ground vehicles is proposed. At present, there are no clear guidelines of how monitoring vehicles must be configured for optimally tracking crop parameters at high resolution. This paper structures the architecture for such vehicles in four subsystems, examines the most common components for each subsystem, and delves into their interactions for an efficient delivery of high-density field data from initial acquisition to final recommendation. Its main advantages rest on the real time generation of crop maps that blend the global positioning of canopy location, some of their agronomical traits, and the precise monitoring of the ambient conditions surrounding such canopies. As a use case, the envisioned architecture was embodied in an autonomous robot to automatically sort two harvesting zones of a commercial vineyard to produce two wines of dissimilar characteristics. The information contained in the maps delivered by the robot may help growers systematically apply differential harvesting, evidencing the suitability of the proposed architecture for massive monitoring and subsequent data-driven actuation. While many crop parameters still cannot be measured non-invasively, the availability of novel sensors is continually growing; to benefit from them, an efficient and trustable sensing architecture becomes indispensable.

Raman-Based Diagnostics of Biotic and Abiotic Stresses in Plants. A Review

Digital farming is a novel agricultural philosophy that aims to maximize a crop yield with the minimal environmental impact. Digital farming requires the development of technologies that can work directly in the field providing information about a plant health. Raman spectroscopy (RS) is an emerging analytical technique that can be used for non-invasive, non-destructive, and confirmatory diagnostics of diseases, as well as the nutrient deficiencies in plants. RS is also capable of probing nutritional content of grains, as well as highly accurate identification plant species and their varieties. This allows for Raman-based phenotyping and digital selection of plants. These pieces of evidence suggest that RS can be used for chemical-free surveillance of plant health directly in the field. High selectivity and specificity of this technique show that RS may transform the agriculture in the US. This review critically discusses the most recent research articles that demonstrate the use of RS in diagnostics of abiotic and abiotic stresses in plants, as well as the identification of plant species and their nutritional analysis.

IoT Sensing Platform as a Driver for Digital Farming in Rural Africa

Small-scale farming can benefit from the usage of information and communication technology (ICT) to improve crop and soil management and increase yield. However, in order to introduce digital farming in rural areas, related ICT solutions must be viable, seamless and easy to use, since most farmers are not acquainted with technology. With that in mind, this paper proposes an Internet of Things (IoT) sensing platform that provides information on the state of the soil and surrounding environment in terms of pH, moisture, texture, colour, air temperature, and light. This platform is coupled with computer vision to further analyze and understand soil characteristics. Moreover, the platform hardware is housed in a specifically designed robust casing to allow easy assembly, transport, and protection from the deployment environment. To achieve requirements of usability and reproducibility, the architecture of the IoT sensing platform is based on low-cost, off-the-shelf hardware and software modularity, following a do-it-yourself approach and supporting further extension. In-lab validations of the platform were carried out to finetune its components, showing the platform’s potential for application in rural areas by introducing digital farming to small-scale farmers, and help them delivering better produce and increasing income.