Machine learning (ML) and deep learning (DL) have achieved great success in different tasks. These include computer vision, image segmentation, natural language processing, predicting classification, evaluating time series, and predicting values based on a series of variables. As artificial intelligence progresses, new techniques are being applied to areas like optical spectroscopy and its uses in specific fields, such as the agrifood industry. The performance of ML and DL techniques generally improves with the amount of data available. However, it is not always possible to obtain all the necessary data for creating a robust dataset. In the particular case of agrifood applications, dataset collection is generally constrained to specific periods. Weather conditions can also reduce the possibility to cover the entire range of classifications with the consequent generation of imbalanced datasets. To address this issue, data augmentation (DA) techniques are employed to expand the dataset by adding slightly modified copies of existing data. This leads to a dataset that includes values from laboratory tests, as well as a collection of synthetic data based on the real data. This review work will present the application of DA techniques to optical spectroscopy datasets obtained from real agrifood industry applications. The reviewed methods will describe the use of simple DA techniques, such as duplicating samples with slight changes, as well as the utilization of more complex algorithms based on deep learning generative adversarial networks (GANs), and semi-supervised generative adversarial networks (SGANs).

Pollution in cities induces harmful effects on human health, which continuously increases the global demand of gas sensors for air quality control and monitoring. In the same manner, the industrial sector requests new gas sensors for their productive processes. Moreover, the association between exhaled gases and a wide range of diseases or health conditions opens the door for new diagnostic applications. The large number of applications for gas sensors has permitted the development of multiple sensing technologies. Among them, optical fiber gas sensors enable their utilization in remote locations, confined spaces or hostile environments as well as corrosive or explosive atmospheres. Particularly, Lossy Mode Resonance (LMR)-based optical fiber sensors employ the traditional metal oxides used for gas sensing purposes for the generation of the resonances. Some research has been conducted on the development of LMR-based optical fiber gas sensors; however, they have not been fully exploited yet and offer optimal possibilities for improvement. This review gives the reader a complete overview of the works focused on the utilization of LMR-based optical fiber sensors for gas sensing applications, summarizing the materials used for the development of these sensors as well as the fabrication procedures and the performance of these devices.

The use of planar waveguides has recently shown great success in the field of optical sensors based on the Lossy Mode Resonance (LMR) phenomenon. The properties of Graphene Oxide (GO) have been widely exploited in various sectors of science and technology, with promising results for gas sensing applications. This work combines both, the LMR-based sensing technology on planar waveguides and the use of a GO thin film as a sensitive coating, to monitor ethanol, water, and acetone. Experimental results on the fabrication and performance of the sensor are presented. The obtained results showed a sensitivity of 3.1, 2.0, and 0.6 pm/ppm for ethanol, water, and acetone respectively, with a linearity factor R² > 0.95 in all cases.

The objective of this Letter consists of the exploration of the lossy mode resonance (LMR) phenomenon beyond the near-infrared region and specifically in the short wave infrared region (SWIR) and medium wave infrared region (MWIR). The experimental and theoretical results show for the first time, to the best of our knowledge, not only LMRs in these regions, but also the utilization of fluoride glass optical fiber associated with this phenomenon. The fabricated devices consist of a nanometric thin-film of titanium dioxide used as LMR generating material, which probed extraordinary sensitivities to external refractive index (RI) variations. RI sensitivity was studied in the SWIR and MWIR under different conditions, such as the LMR wavelength range or the order of resonance, showing a tremendous potential for the detection of minute concentrations of gaseous or biological compounds in different media.

This work presents a first approach to the utilization of Lossy Mode Resonance (LMR) based optical fiber sensors for gas detection. The optical sensor is based on a SnO₂ thin-film fabricated onto the core of cladding removed multimode fibers (MMF). The time response of the device to four different gases (NH₃, NO, CO₂ and O₂) was monitored obtaining the best sensitivity for NO whereas the response to NH₃ revealed the best repeatability.