Proposal of a New System for Essential Oil Classification Based on Low-Cost Gas Sensor and Machine Learning Techniques.

Essential oils are valuable in various industries, but their easy adulteration can cause adverse health effects. Electronic nasal sensors offer a solution for adulteration detection. This article proposes a new system for characterising essential oils based on low-cost sensor networks and machine learning techniques. The sensors used belong to the MQ family (MQ-2, MQ-3, MQ-4, MQ-5, MQ-6, MQ-7, and MQ-8). Six essential oils were used, including Cistus ladanifer, Pinus pinaster, and Cistus ladanifer oil adulterated with Pinus pinaster, Melaleuca alternifolia, tea tree, and red fruits. A total of up to 7100 measurements were included, with more than 118 h of measurements of 33 different parameters. These data were used to train and compare five machine learning algorithms: discriminant analysis, support vector machine, k-nearest neighbours, neural network, and naive Bayesian when the data were used individually or when hourly mean values were included. To evaluate the performance of the included machine learning algorithms, accuracy, precision, recall, and F1-score were considered. The study found that using k-nearest neighbours, accuracy, recall, F1-score, and precision values were 1, 0.99, 0.99, and 1, respectively. The accuracy reached 100% with k-nearest neighbours using only 2 parameters for averaged data or 15 parameters for individual data.

Low-Cost Microcontroller-Based Multiparametric Probe for Coastal Area Monitoring.

The monitoring of the coastal environment is a crucial factor in ensuring its proper management. Nevertheless, existing monitoring technologies are limited due to their cost, temporal resolution, and maintenance needs. Therefore, limited data are available for coastal environments. In this paper, we present a low-cost multiparametric probe that can be deployed in coastal areas and integrated into a wireless sensor network to send data to a database. The multiparametric probe is composed of physical sensors capable of measuring water temperature, salinity, and total suspended solids (TSS). The node can store the data in an SD card or send them. A real-time clock is used to tag the data and to ensure data gathering every hour, putting the node in deep sleep mode in the meantime. The physical sensors for salinity and TSS are created for this probe and calibrated. The calibration results indicate that no effect of temperature is found for both sensors and no interference of salinity in the measuring of TSS or vice versa. The obtained calibration model for salinity is characterised by a correlation coefficient of 0.9 and a Mean Absolute Error (MAE) of 0.74 g/L. Meanwhile, different calibration models for TSS were obtained based on using different light wavelengths. The best case was using a simple regression model with blue light. The model is characterised by a correlation coefficient of 0.99 and an MAE of 12 mg/L. When both infrared and blue light are used to prevent the effect of different particle sizes, the determination coefficient of 0.98 and an MAE of 57 mg/L characterised the multiple regression model.

Incorrect Facemask-Wearing Detection Using Convolutional Neural Networks with Transfer Learning.

The COVID-19 pandemic has been a worldwide catastrophe. Its impact, not only economically, but also socially and in terms of human lives, was unexpected. Each of the many mechanisms to fight the contagiousness of the illness has been proven to be extremely important. One of the most important mechanisms is the use of facemasks. However, the wearing the facemasks incorrectly makes this prevention method useless. Artificial Intelligence (AI) and especially facial recognition techniques can be used to detect misuses and reduce virus transmission, especially indoors. In this paper, we present an intelligent method to automatically detect when facemasks are being worn incorrectly in real-time scenarios. Our proposal uses Convolutional Neural Networks (CNN) with transfer learning to detect not only if a mask is used or not, but also other errors that are usually not taken into account but that may contribute to the virus spreading. The main problem that we have detected is that there is currently no training set for this task. It is for this reason that we have requested the participation of citizens by taking different selfies through an app and placing the mask in different positions. Thus, we have been able to solve this problem. The results show that the accuracy achieved with transfer learning slightly improves the accuracy achieved with convolutional neural networks. Finally, we have also developed an Android-app demo that validates the proposal in real scenarios.