Search Results (9)

Demand and market value for pennywort largely depend on the quality of the leaves, which can be affected by various ambient environment or fertigation variables during cultivation. Although early detection of defects in pennywort leaves would enable growers to take quick action, conventional manual detection is laborious and time consuming as well as subjective. Therefore, the objective of this study was to develop an automatic leaf defect detection algorithm for pennywort plants grown under controlled environment conditions, using machine vision and deep learning techniques. Leaf images were captured from pennywort plants grown in an ebb-and-flow hydroponic system under fluorescent light conditions in a controlled plant factory environment. Physically or biologically damaged leaves (e.g., curled, creased, discolored, misshapen, or brown spotted) were classified as defective leaves. Images were annotated using an online tool, and Mask R-CNN models were implemented with the integrated attention mechanisms, convolutional block attention module (CBAM) and coordinate attention (CA) and compared for improved image feature extraction. Transfer learning was employed to train the model with a smaller dataset, effectively reducing processing time. The improved models demonstrated significant advancements in accuracy and precision, with the CA-augmented model achieving the highest metrics, including a mean average precision (mAP) of 0.931 and an accuracy of 0.937. These enhancements enabled more precise localization and classification of leaf defects, outperforming the baseline Mask R-CNN model in complex visual recognition tasks. The final model was robust, effectively distinguishing defective leaves in challenging scenarios, making it highly suitable for applications in precision agriculture. Future research can build on this modeling framework, exploring additional variables to identify specific leaf abnormalities at earlier growth stages, which is crucial for production quality assurance. Full article

This study aimed to simulate dynamic irrigation management zones (MZs) in two maize fields for a variable rate hose reel fertigation machine (VRFM) with a four-section boom control. Soil moisture content was measured from nine and four soil moisture sensors in Field 1 (8.2 ha) and Field 2 (2.5 ha), respectively, on different dates during the 2022 crop season. Three and five MZs scenarios were simulated per irrigation and the theoretical maps were processed for implementation. The application maps fitted to the VRFM showed significant spatiotemporal variations in irrigation requirements. For instance, in Field 1, 3-MZ modelling showed that the areas requiring high (H), medium (M), and low (L)-level irrigation on 21 July were 1.60, 4.84, and 1.85 ha, respectively, even though the farmer applied uniform rate over the whole field. H-level sub-areas ranged between 1.22 ha (25 July) and 3.25 ha (7 July), showing a coefficient of variation (CV) of 43.32% for the three MZs, whereas H-level sub-areas for the five MZs varied from 0.41 ha (2 July) to 1.49 ha (7 July) with a CV value of 48.84%. High levels of within-field variability can be addressed using precise and dynamic irrigation MZs fitted to the irrigation technology used. Full article

In addressing the control of asynchronous motors in the remote conductivity of fertigation machines, this study proposes a joint control strategy based on the Fast Terminal Sliding Mode Control-Disturbance Observer (FTSMC-DO) system for asynchronous motors. The goal is to enhance the dynamic performance and disturbance resistance of asynchronous motors, particularly under low-speed operating conditions. The approach involves refining the two-degree-of-freedom internal model controller using fractional-order functions to explicitly separate the controller’s robustness and tracking capabilities. To mitigate the motor’s sensitivity to external disturbances during variable speed operations, a load disturbance observer is introduced, employing hyperbolic tangent and Fal functions for real-time monitoring and compensation, seamlessly integrated into the sliding mode controller. To address issues related to low-speed chattering typically associated with sliding mode controllers, this study introduces a revised non-singular fast terminal sliding mode surface. Additionally, guided by fuzzy control principles, the study enables real-time selection of sliding mode approaching law parameters. Experimental results from the asynchronous motor control platform demonstrate that FTSMC-DO control significantly reduces adjustment time and speed fluctuations during operation, minimizing the impact of load disturbances on the system. The system exhibits robust disturbance rejection, improved robustness, and enhanced control capability. Furthermore, field tests validate the effectiveness of the FTSMC-DO system in regulating remote electrical conductivity (EC) levels. The control time is observed to be less than 120 s, overshoot less than 16.1%, and EC regulation within 0.2 mS·cm⁻¹ over a pipeline distance of 120 m. The FTSMC-DO control consistently achieves the desired EC levels with minimal fluctuation and overshoot, outperforming traditional PID and SMC methods. This high level of precision is crucial for ensuring optimal nutrient delivery and efficient water usage in agricultural irrigation systems, highlighting the system’s potential as a valuable tool in modern, sustainable farming practices. Full article

Greenhouse climate control systems are usually based on greenhouse microclimate settings to exert any control. However, to save energy, water and nutrients, additional parameters related to crop performance and physiology will have to be considered. In addition, detecting crop stress before it is clearly visible by naked eye is an advantage that could aid in microclimate control. In this study, a Machine Learning (ML) model which takes into account microclimate and crop physiological data to detect different types of crop stress was developed and tested. For this purpose, a multi-sensor platform was used to record tomato plant physiological characteristics under different fertigation and air temperature conditions. The innovation of the current model lies in the integration of photosynthesis rate (Ps) values estimated by means of remote sensing using a photochemical reflectance index (PRI). Through this process, the time-series Ps data were combined with crop leaf temperature and microclimate data by means of the ML model. Two different algorithms were evaluated: Gradient Boosting (GB) and MultiLayer perceptron (MLP). Two runs with different structures took place for each algorithm. In RUN 1, there were more feature inputs than the outputs to build a model with high predictive accuracy. However, in order to simplify the process and develop a user-friendly approach, a second, different run was carried out. Thus, in RUN 2, the inputs were fewer than the outputs, and that is why the performance of the model in this case was lower than in the case of RUN 1. Particularly, MLP showed 91% and 83% accuracy in the training sample, and 89% and 82% in testing sample, for RUNs 1 and 2, respectively. GB showed 100% accuracy in the training sample for both runs, and 91% and 83% in testing sample in RUN 1 and RUN 2, respectively. To improve the accuracy of RUN 2, a larger database is required. Both models, however, could easily be incorporated into existing greenhouse climate monitoring and control systems, replacing human experience in detecting greenhouse crop stress conditions. Full article

Urban agriculture can be shortly defined as the growing of plants and/or the livestock husbandry in and around cities. Although it has been a common occupation for the urban population all along, recently there is a growing interest in it both from public bodies and researchers, as well as from ordinary citizens who want to engage in self-cultivation. The modern citizen, though, will hardly find the free time to grow his own vegetables as it is a process that requires, in addition to knowledge and disposition, consistency. Given the above considerations, the purpose of this work was to develop an economic robotic system for the automatic monitoring and management of an urban garden. The robotic system was designed and built entirely from scratch. It had to have suitable dimensions so that it could be placed in a balcony or a terrace, and be able to scout vegetables from planting to harvest and primarily conduct precision irrigation based on the growth stage of each plant. Fertigation and weed control will also follow. For its development, a number of technologies were combined, such as Cartesian robots’ motion, machine vision, deep learning for the identification and detection of plants, irrigation dosage and scheduling based on plants’ growth stage, and cloud storage. The complete process of software and hardware development to a robust robotic platform is described in detail in the respective sections. The experimental procedure was performed for lettuce plants, with the robotic system providing precise movement of its actuator and applying precision irrigation based on the specific needs of the plants. Full article

Demand for agricultural products is increasing as population continues to grow in Africa. To attain a higher crop yield while preserving the environment, appropriate management of macronutrients (i.e., nitrogen (N), phosphorus (P) and potassium (K)) and crops are of critical prominence. This paper aims to review the state of art of the use of remote sensing in soil agricultural applications, especially in monitoring NPK availability for widely grown crops in Africa. In this study, we conducted a substantial literature review of the use of airborne imaging technology (e.g., different platforms and sensors), methods available for processing and analyzing spectral information, and advances of these applications in farming practices by the African scientific community. Here we aimed to identify knowledge gaps in this field and challenges related to the acquisition, processing, and analysis of hyperspectral imagery for soil agriculture investigations. To do so, publications over the past 10 years (i.e., 2008–2021) in hyperspectral imaging technology and applications in monitoring macronutrients status for crops were reviewed. In this study, the imaging platforms and sensors, as well as the different methods of processing encountered across the literature, were investigated and their benefit for NPK assessment were highlighted. Furthermore, we identified and selected particular spectral regions, bands, or features that are most sensitive to describe NPK content (both in crop and soil) that allowed to characterize NPK. In this review, we proposed a hyperspectral data-based research protocol to quantify variability of NPK in soil and crop at the field scale for the sake of optimizing fertilizers application. We believe that this review will contribute promoting the adoption of hyperspectral technology (i.e., imaging and spectroscopy) for the optimization of soil NPK investigation, mapping, and monitoring in many African countries. Full article

K (K) is of paramount importance for apple (Malus × domestica Borkh.), not only for tree growth and development but also for the size and quality of fruit yield. The apple plant’s demand for K varies, along with the progression of phenological phases, during the growing season. The K demand peaks during ripening of fruits featuring relatively high concentration of K comparable to that of the leaves. The mainstream method of apple tree K fertilization is through application of the fertilizer to the soils to improve K uptake by the roots. The bioavailability of K depends on assorted various factors, including pH, interaction with other nutrients in soil solution, temperature, and humidity. An important role in making the K from soil available for uptake by plants is played by plant growth-promoting microorganisms (PGPM), and the specific role of the PGPM is discussed. Advantages of fertigation (the combination of irrigation and fertilization) as an approach include allowing to balance application rate of K fertilizer against its variable demand by plants during the growing season. Excess K in the soil leads to competitive inhibition of calcium uptake by plants. The K-dependent deficiency of Ca leads to its predominant channeling to the leaves and hence to its decline in fruits. Consequently, the apple fruits affected by the K/Ca imbalance frequently develop physiological disorders in storage. This emphasizes the importance of the balanced K application, especially during the last months of the growing season, depending on the crop load and the actual K demand. The potential use of modern approaches to automated crop load estimation through machine vision for adjustment of K fertilization is underlined. Full article

Fertigation management of banana plantations at a plot scale is expanding rapidly in Brazil. To guide nutrient management at such a small scale, genetic, environmental and managerial features should be well understood. Machine learning and compositional data analysis (CoDa) methods can measure the effects of feature combinations on banana yield and rank nutrients in the order of their limitation. Our objectives are to review ML and CoDa models for application at regional and local scales, and to customize nutrient diagnoses of fertigated banana at the plot scale. We documented 940 “Prata” and “Cavendish” plot units for tissue and soil tests, environmental and managerial features, and fruit yield. A Neural Network informed by soil tests, tissue tests and other features was the most proficient learner (AUC up to 0.827). Tissue nutrients were shown to have the greatest impact on model accuracy. Regional nutrient standards were elaborated as centered log ratio means and standard deviations of high-yield and nutritionally balanced specimens. Plot-scale diagnosis was customized using the closest successful factor-specific tissue compositions identified by the smallest Euclidean distance from the diagnosed composition using centered or isometric log ratios. Nutrient imbalance differed between regional and plot-scale diagnoses, indicating the profound influence of local factors on plant nutrition. However, plot-scale diagnoses require large, reliable datasets to customize nutrient management using ML and CoDa models. Full article

Fertigation is a widely used crop growing method that consists on the precise injection of a nutrient solution that commonly consists on a mixture of three basis components (nitrogen, phosphorus and potassium) diluted in water. This nutritive suspension is delivered to the plants with a frequency and relative basis contents that depends on the plant’s type, its vegetative state and actual environmental conditions. This production process is becoming increasingly popular due to several advantages over more traditional approaches such as more control on the plant fertilisers and an increasing reduction of irrigation water. This is achieved by an increase complexity on the crop growing process management which requires a technological layer responsible for mixing the nutrients and monitoring the local environmental conditions. Despite this technical component, the short and long term management decisions depend almost exclusively on the grower’s experience and intuition. This type of human-on-the-loop control can lead to a suboptimal use of resources wish can translate into reduction of economic profit and an can lead to waste of water and fertilisers. In this context, decision support mechanisms based on artificial intelligence and machine learning algorithms can be of extreme relevance in order to steer the grower decisions and increase the overall production process efficiency. The performance of those types of approaches strongly relies on the availability of data which can be both local and global. This work deals with the architecture of a sensor network which will be responsible to gather local information on the actual growing conditions. Those conditions are usually not homogeneous within the complete production plant and must be taken into consideration. In particular, the current architecture vision will consider those clusters, where the environmental conditions are similar, as cyberphysical devices. These devices will consist on vegetative production area, sensor networks and local control of irrigation state. Full article