Accurate classification of tomato ripening stages and quality estimation is pivotal for optimizing post-harvest management and ensuring market value. This study presents a rigorous comparative analysis of morphological and colorimetric features extracted via two state-of-the-art deep learning-based instance segmentation frameworks—Mask R-CNN and YOLOv8n-seg—and their efficacy in machine learning-driven ripening stage classification and quality prediction. Using 216 fresh-market tomato fruits across four defined ripening stages, we extracted 27 image-derived features per model, alongside 12 laboratory-measured physio-morphological traits. Multivariate analyses revealed that R-CNN features capture nuanced colorimetric and structural variations, while YOLOv8 emphasizes morphological characteristics. Machine learning classifiers trained with stratified 10-fold cross-validation achieved up to 95.3% F1-score when combining both feature sets, with R-CNN and YOLOv8 alone attaining 96.9% and 90.8% accuracy, respectively. These findings highlight a trade-off between the superior precision of R-CNN and the real-time scalability of YOLOv8. Our results demonstrate the potential of integrating complementary segmentation-derived features with laboratory metrics to enable robust, non-destructive phenotyping. This work advances the application of vision-based machine learning in precision agriculture, facilitating automated, scalable, and accurate monitoring of fruit maturity and quality.

In recent years, the use of warm-season species, which are species requiring less water, has been pursued in continental areas, but their dormancy and spring green-up need to be properly defined. In urban green areas, we find that small-scale microclimatic differences, while less intense than classical urban–rural gradients, still influence vegetation performance and spring green-up. This study examines the impact of microclimatic temperature variation on the spring green-up of different cool-season and warm-season turfgrasses in the continental climate of Madrid, Spain. The evaluation of colour change during the spring green-up process has been conducted using different vegetation indices, and mathematical models for correlating temperature with the indices’ values have been obtained. The results indicate that with average temperatures varying by about 1.3 °C and 0.9 °C in January and February, respectively, there have been marked differences in spring green-up, especially in cool-season turfgrasses, of almost one month. In contrast, differences in warm-season turfgrasses were reduced. Among the four vegetation indices, Canopeo has proved to be the best for detecting the early stages of spring green-up, with R2 values ranging from 0.43 to 0.92. Meanwhile, the tailored greenness index for turfgrass was the most effective for determining the moment at which warm-season grasses achieve the colouration of cool-season grasses, with R2 ranging from 0.79 to 0.85. Finally, the green leaf index was particularly valuable for identifying differences among species and sectors throughout the entire spring green-up process. Models based on this index achieve high R2 values (0.57 to 0.94), but these models predict the moment at which warm-season grasses achieve cool-season grasses’ colouration later than it actually occurs. Understanding how turfgrasses respond to these localised microclimatic conditions is essential for selecting resilient species and improving maintenance strategies in parks, sports areas, and other components of urban green infrastructure.

Volatile organic compounds (VOCs) are central to the aromatic and therapeutic properties of essential oils (EOs), with their profiles serving as reliable indicators of EO quality. In Cistus ladanifer, the synthesis and emission of VOCs—particularly terpenic hydrocarbons—are strongly influenced by environmental variables such as temperature, humidity, and solar radiation. Traditional EO quality assessment methods, including gas chromatography (GC), although highly accurate, are costly, labor-intensive, and destructive. This study proposes a smart sensor system that utilizes a low-cost array of MQ gas sensors combined with machine learning (ML) algorithms for non-destructive classification of VOC fingerprints in Cistus ladanifer EO. VOC data from 33 EO samples were collected using MQ sensors and paired with environmental datasets (Daily, 15-Day, and All) obtained from weather stations in the cultivation areas. Gas chromatography coupled with flame ionization detection and mass spectrometry (GC-FID/MS) was used as the reference method to quantify the concentrations of terpenic hydrocarbons. A total of 5,154 data points were used, with 75

Water quality is being altered by human activities and monitoring them is highly recommended. Nevertheless, low-cost monitoring nodes are generally limited in terms of computation capacity. In this paper, we analyse the performance of different machine learning algorithms trained and tested in nodes as part of edge computing in a water quality monitoring network. The performance of five well-known algorithms for data classification in multiclass and binary classification is assessed. The algorithms are coded to run in a low-cost node, and their performance is compared with Matlab results. Accuracy and precision are the metrics used for the comparison, with the artificial neural network as the algorithm with higher performance, followed by the Naive Bayesian classifier. The accuracy for the multiclass classification in the node is slightly better, 98.66

The quality of hydrobiological species for human consumption, such as fish and shellfish, is crucial to ensure their safety. This article highlights the use of Metal Oxide Semiconductor Resistive (MOSR)-based MQ sensor modules, along with temperature and humidity, to analyze these species. The prototype is based on the use of an Arduino Mega 2560 Rev3 to connect the sensors with data stored on a Raspberry Pi 4, which can be accessed via Wi-Fi for downloading. Machine learning (ML) models were applied to classify and estimate non-refrigeration time with different quantities of variables, using species such as Illex argentinus, Sardina pilchardus, Scomber colias, and Sepia pharaonis. In total, 33 classification models and 28 regression models were employed. The best results were obtained with Boosted Trees model, achieving an accuracy of 99.93% in the validation phase and 99.74% in the testing phase. In the regression models to estimate non-refrigeration time, with 38 variables, the best model was Bagged Trees, with an R-squared of 0.995 for the validation phase and 0.997 for the testing phase.

In coastal areas, saltwater and freshwater converge, creating high biodiversity and significant pollution issues from industrial waste, microplastics, fertilizers, and medications. These substances disrupt ecosystems and species' life cycles that regulate key physiological processes, even affecting humans. Chromatography is the most commonly used technique for detecting drugs in water, but it is also expensive and requires specialized personnel. This study used optical sensors around a tube to detect drugs (Dacortin, Diazepam, and Prednisone) and baking flour in water by measuring light absorption at different wavelengths (visible, infrared, and ultraviolet). Data changes were processed using an Arduino Uno microprocessor, configured as an IoT device. The concentrations were 0, 10, 15, and 30 mg/100 mL. The results display several models using MATLAB. K-nearest neighbor showed the best performance with 100% accuracy and the shortest training time (0.71 seconds). Ultraviolet sensors provided the most variable and impactful data, while red and green sensors showed similar results, and blue had higher dispersion. Infrared sensors were less sensitive, requiring future adjustment for improved analysis. Regression models showed high performance, the best being Rational Quadratic Gaussian Process Regression, achieving a Root Mean Squared Error of 0.2 mg/100 mL in training and 0.01 mg/100 mL in testing. Some training errors were attributed to outliers caused by electronic or sample issues, highlighting the need for outlier filtering. Test dataset errors were minimal and consistent with training results, validating the model's accuracy.

Wireless multimedia sensor networks (WMSN) have been mainly applied in urban areas using Zigbee, Bluetooth or Wi-Fi to forward the data, but no application has been designed in rural areas due to the large distances and energy restrictions. The related literature shows only one deployment with LoRa using a Raspberry Pi 3 node, but no measure of real video delivery was performed. In this paper, we propose the use of LoRa technology for video streaming in a WMSN for smart agriculture. The objective of this design is to monitor the fields by having nodes with sensors, nodes with cameras, and aggregator nodes. Two nodes are used to test the performance of the network and ensure that video streaming with LoRa is possible. The emitter node, an Arduino ATMega, sends a video to the receiver node, an ESP32. The LoRa UART SX1278 was used in both nodes to communicate them. Results in the laboratory, with both nodes close, indicate that the minimum latency is achieved using a block size of 6408 bytes. Four video resolutions and 4 video extensions are tested to evaluate the maximum video quality which can be streamed. The results in a real scenario with nodes spaced 731 meters indicated that video in .mp4 video format at 240 p resolution can be streamed correctly with a medium access time of 3.2 s and 0% of packet loss for a video of 5 s. Thus, it is possible to affirm that LoRa can be used for video streaming.

Monitoring ocean salinity and temperature is vital for understanding ecosystem health and mitigating marine pollution from untreated municipal sewage. This paper explores the use of IoT systems for continuous underwater monitoring of marine outfalls from wastewater treatment plants, providing early warnings of contamination. The system is composed of multiple sensors, highlighting the temperature and salinity sensors in this paper. We have used a commercial temperature sensor, the DS18B20 and a tailored salinity sensor. Those sensors are connected to the sensor node, an ESP32. The node collects the data and sends it to a database. The temperature and salinity sensors are calibrated and validated with generated samples. The samples have a variation of salinity from 0 to 40 g/L, and the variation in temperature goes from 40 to 10 degrees C. The results indicated that the commercial temperature sensor has a bias, and we proposed an equation to correct it. The salinity sensor is calibrated, obtaining an equation with an R2 of 99.5 %. The network traffic analysis indicates an increase in traffic when errors in TPC packets appear.

Soil degradation issues, such as salinization, waterlogging, and erosion, pose significant challenges in agricultural irrigation and fertigation. Conventional laboratory methods for analyzing soil conditions are often cost-prohibitive, time-intensive, and require specialized personnel. This study investigates the application of a virtual sensor, incorporating RGB and LDR modules, to measure fertilizer concentration and water content in soil, as well as to differentiate between water and fertilizer. Statistical analysis of the acquired data using ANOVA revealed significant differences between samples under blue, yellow, and magenta lights, with p-values of 0.0425, 0.018, and 0.0425, respectively. Machine Learning (ML) techniques were employed to enhance the system's performance, yielding an 87.5% accuracy in binary classification using the K-Nearest Neighbors (KNN) and Euclidean models. Additionally, regression models demonstrated R2 values of 0.802 and 0.891 for irrigated and ferti-irrigated soils, respectively. These results suggest that the proposed optical sensor system is a promising, cost-effective alternative to traditional methods, though further validation in real-world applications is necessary.

Turbidity is one of the crucial parameters of water quality. Even though many commercial devices, low-cost sensors, and remote sensing data can efficiently quantify turbidity, they are not valid tools for the classification it. In this paper, we design, calibrate, and test a novel optical low-cost sensor for turbidity quantification and classification. The sensor is based on an RGB light source and a light detector. The analyzed samples are characterized by turbidity values from 0.02 to 60 NTUs, and have four different sources. These samples were generated to represent natural turbidity sources and leaves in the marine areas close to agricultural lands. The data are gathered using 64 different combinations of light, generating complex matrix data. Machine learning models are compared to analyze this data, including training, validation, and test datasets. Moreover, different alternatives for data preprocessing and feature selection are assessed. Concerning the quantification of turbidity, the best results were obtained using averaged data and principal components analyses in conjunction with exponential gaussian process regression, achieving an R2 of 0.979. Regarding the classification of the turbidity, an accuracy of 91.23% is obtained with the fine K-Nearest-Neighbor classifier. The cases in which data were misclassified are characterized by turbidity values lower than 5 NTUs. The obtained results represent an improvement over the current solutions in terms of turbidity quantification and a completely novel approach to turbidity classification.

It is known that natural products can be used to strengthen and minimise stress of the gardening and sportive lawns, thus reducing the required inputs. In this paper, a trial is designed that allows for the study of the effect of a combination of two biostimulants and water -retaining agent products on different types of lawns. During 6 months, including the summer, soil and plant parameters are evaluated to compare the effects of treatments on soil temperature, humidity, and electrical conductivity, along with the NDVI of the grasslands. Treatment with the water -retaining agent and the second tested biostimulants has increased soil moisture by 10 %, with a greater effect on ornamental grasslands with lower maintenance requirements than sports lawns. The treatments with the two biostimulants without the water retaining agent do not lead to a significant variation in the aspect of the lawn. Marginal increases in the NDVI have been observed in all the treatments, which include the biostimulants. According to these results, it is possible to achieve better water efficiency in managing urban lawns by using natural products, which leads to a more sustainable use of hydric resources.

Drones are being used for agriculture monitoring in many different crops. Nevertheless, the use of drones for green areas’ evaluation is limited, and information is scattered. In this survey, we focus on the collection and evaluation of existing experiences of using drones for turfgrass monitoring. Despite a large number of initial search results, after filtering the information, very few papers have been found that report the use of drones in green areas. Several aspects of drone use, the monitored areas, and the additional ground-based devices for information monitoring are compared and evaluated. The data obtained are first analysed in a general way and then divided into three groups of papers according to their application: irrigation, fertilisation, and others. The main results of this paper indicate that despite the diversity of drones on the market, most of the researchers are using the same drone. Two options for using cameras in order to obtain infrared information were identified. Moreover, differences in the way that drones are used for monitoring turfgrass depending on the aspect of the area being monitored have been identified. Finally, we have indicated the current gaps in order to provide a comprehensive view of the existing situation and elucidate future trends of drone use in turfgrass management.

Studying soil composition is vital for agricultural and edaphology disciplines. Presently, colorimetry serves as a prevalent method for the on-site visual examination of soil characteristics. However, this technique necessitates the laboratory-based analysis of extracted soil fragments by skilled personnel, leading to substantial time and resource consumption. Contrastingly, sensor techniques effectively gather environmental data, though they mostly lack in situ studies. Despite this, sensors offer substantial on-site data generation potential in a non-invasive manner and can be included in wireless sensor networks. Therefore, the aim of the paper is to develop a low-cost red, green, and blue (RGB)-based sensor system capable of detecting changes in the composition of the soil. The proposed sensor system was found to be effective when the sample materials, including salt, sand, and nitro phosphate, were determined under eight different RGB lights. Statistical analyses showed that each material could be classified with significant differences based on specific light variations. The results from a discriminant analysis documented the 100% prediction accuracy of the system. In order to use the minimum number of colors, all the possible color combinations were evaluated. Consequently, a combination of six colors for salt and nitro phosphate successfully classified the materials, whereas all the eight colors were found to be effective for classifying sand samples. The proposed low-cost RGB sensor system provides an economically viable and easily accessible solution for soil classification.

Projections indicate aquaculture will produce 106 million tonnes of fish by 2030, emphasizing the need for efficient and sustainable practices. New technologies can provide a valuable tool for adequate fish farm management. The aim of this paper is to explore the factors affecting fish well-being, the design of control systems for aquaculture, and the proposal of a smart system based on algorithms to improve efficiency and sustainability. First, we identify the domains affecting fish well-being: the production domain, abiotic domain, biotic domain, and control systems domain. Then, we evaluate the interactions between elements present in each domain to evaluate the key aspects to be monitored. This is conducted for two types of fish farming facilities: cages in the sea and recirculating aquaculture systems. A total of 86 factors have been identified, of which 17 and 32 were selected to be included in monitoring systems for sea cages and recirculating aquaculture systems. Then, a series of algorithms are proposed to optimize fish farming management. We have included predefined control algorithms, energy-efficient algorithms, fault tolerance algorithms, data management algorithms, and a smart control algorithm. The smart control algorithms have been proposed considering all the aforementioned factors, and two scenarios are simulated to evaluate the benefits of the smart control algorithm. In the simulated case, the turbidity when the control algorithm is used represents 12.5% of the turbidity when not used. Their use resulted in a 35% reduction in the energy consumption of the aerator system when the smart control was implemented.

The ocean, with its intricate processes, plays a pivotal role in shaping marine life, habitats, and the Earth’s climate. This study addresses issues such as beach erosion, the survival of propagules from species like Posidonia oceanica, and nutrient distribution. To tackle these challenges, we propose an innovative sensor that quantifies hydrodynamic velocity by measuring the output voltage derived from detecting changes in light absorption and scattering using LEDs and LDRs. Our results not only demonstrate the effectiveness of the sensor but also the accuracy of the processing algorithm. Notably, the blue LED exhibited the lowest mean relative error of 7.59% in freshwater, while the yellow LED was most precise in chlorophyll-containing water, with a mean relative error of 6.80%. In a runoff simulation, we observed similar velocities with the blue, green, and white LEDs, 6.89 cm/s, 6.99 cm/s, and 7.05 cm/s, respectively, for nearly identical time intervals. It is important to highlight that our proposed sensor is not only effective but also highly cost-efficient, representing less than 0.43% of the cost of a Nortek Vector 6 MHz and 0.18% of the Teledyne Workhorse II 300 kHz Marine. This makes it a key tool for managing marine ecosystems sustainably.

Urban air quality, impacted by human-made pollution, impacts health and requires continuous monitoring. MQ sensors are the preferred air quality sensors despite their high energy consumption due to their cost, requiring the use machine learning to classify different types of air. The aim of this article is to evaluate a monitoring solution with low-cost and low-energy consumption to classify urban and rural air. A single MQ sensor will be used with a network with edge and fog computing to balance the energy consumption. Edge computing was included in the node for feature extraction, and fog computing was applied in the smartphone to classify the data using machine learning. Different sensors and time buffers are compared in order to find the adequate sensor for data generation and time buffer for feature extraction. The results indicate that it has been possible to achieve accuracies of 100% using a single sensor, the MQ2, with time buffers of 45-60 measures. With this proposal, it is possible to reduce the energy consumed by data gathering to 25% of the original consumption due to the use of a single sensor, due to the reduction in the sensors used in the previous prototype. Moreover, it has been possible to reduce the energy linked to data forwarding by almost 97% due to using a time buffer.

The chemical composition of essential oils (EOs) from Cistus ladanifer has a huge variability throughout the year, impacting the oil quality. Nowadays, EO analytic chemistry techniques, which are expensive and destroy the sample, are utilized to measure the chemical composition. In the paper, we propose a combination of low-cost sensors and machine learning based system. As low-cost sensors, seven gas sensors are combined to obtain up to 36 features. Regarding machine learning, 31 multiclass classification algorithms are applied. Data from sensors were collected for 33 samples of EO from Cistus ladanifer. The generated dataset was split into training and test datasets, with 75 % of the data for training. The datasets were created to ensure a homogeneous chemical composition distribution on both training and test datasets. There were three target chemical compounds: Alpha-pinene and Viridiflorol as individual compounds and Terpenic Hydrocarbons as a group of chemical compounds. The value of the percentage of each targeted compound is converted into a categoric variable with 5 possible values, 1 being the lowest concentration and 5 being the maximum one. The data of the MQ-sensors were included as the input for the models, and each one of the targeted chemical compounds was selected as an output for different models. The input features were ranged using different algorithms for the feature selection process. The results indicate that there is no valid classification model for Viridiflorol, and limited accuracy is achieved for Alpha-pinene. Meanwhile, for Terpenic Hydrocarbons, an accuracy of 91.6 % is achieved. It is important to highlight that these accuracies were attained when a reduced number of features were included, ranging the number of features from 11 to 13. This is the first case in which MQ-based gas sensors, or other metal oxide sensors, are used to correctly determine the concentration of a chemical compounds in a complex matrix formed by dozens of compounds. This system will provide a cheap method to determine the quality of EOs and confirm the benefits of combining low-cost sensors with machine learning.

Smart agricultural solutions contemplate the use of Wireless Sensor Networks (WSNs) to optimize resource use and decision-making. This study proposes deploying a LoRa-based WSN to monitor alpha-pinene production in Cistus ladanifer shrubs, aiming to estimate the optimal harvest time for obtaining the maximum yield of its essential oil. The communication system integrates LoRa and IEEE 802.11 technologies within an IoT framework, utilizing a layered data transmission system comprising edge and fog layers. Data transmission was tested at distances ranging from 678 to 14,700 meters over 28 iterations. MQ gas sensors recorded a-pinene data, achieving a validation dataset accuracy of 99.79% with Cubic SVM and Fine KNN, and a test dataset accuracy of 62.5% with Kernel Naive Bayes using 11 to 13 features. Results showed the implementation of LoRa technology in our system offers substantial benefits in terms of range, reliability, and power efficiency, thus supporting the overall functionality and scalability of the network. Identifying the peak concentration of a-pinene will aid in harvesting Cistus ladanifer at the optimal time to maximize yield. Integration of decision support systems for optimized crop yields and exploring alternative machine learning techniques with higher accuracies could be explored in future studies.

Essential oils are a valuable raw material for several industries. Low-cost methods cannot detect its adulteration; specialised equipment is required. In this paper, we proposed the use of gas sensors to detect the adulteration process in the essential oil of Cistus ladanifer. Gas sensors are used in a measuring chamber to measure pure and adulterated oils. We compare the suitability of the tested sensors for detecting adulterated oil and the required measuring time. A total of five samples are determined, with a measuring time of 12 h. Each gas sensor is configured to be sensitive to different compounds. Even though sensors are not specific to detect the volatile organic compounds (VOCs) present in the essential oil, our objective is to evaluate if these VOCs might interact with the sensors as an interferent. Results indicate that various gas sensors sensitive to the same chemical compound offered different values. It might indicate that the interaction of VOCs is different among the tested sensors or that the location of the sensors and the heterogeneous distribution of VOCs along the measurement chamber impact the data. Regarding the performed analyses, we can affirm that identifying the adulterated essential oil is possible using the generated data. Moreover, the results suggest that most of the data, even for different compounds and sensors, are highly correlated, allowing a reduction in the studied variables. According to the high correlation, data are reduced, and 100% of correct classification can be obtained even when only the MQ3 and MQ8 are used.