A wrapper framework for feature selection and ELM weights optimization for FMG-based sign recognition.

Force myography (FMG) is increasingly gaining importance in gesture recognition because of it's ability to achieve high classification accuracy without having a direct contact with the skin. In this study, we investigate the performance of a bracelet with only six commercial force sensitive resistors (FSR) sensors for classifying many hand gestures representing all letters and numbers from 0 to 10 in the American sign language. For this, we introduce an optimized feature selection in combination with the Extreme Learning Machine (ELM) as a classifier by investigating three swarm intelligence algorithms, which are the binary grey wolf optimizer (BGWO), binary grasshopper optimizer (BGOA), and binary hybrid grey wolf particle swarm optimizer (BGWOPSO), which is used as an optimization method for ELM for the first time in this study. The findings reveal that the BGWOPSO, in which PSO supports the GWO optimizer by controlling its exploration and exploitation using inertia constant to improve the convergence speed to reach the best global optima, outperformed the other investigated algorithms. In addition, the results show that optimizing ELM with BGWOPSO for feature selection can efficiently improve the performance of ELM to enhance the classification accuracy from 32% to 69.84% for classifying 37 gestures collected from multiple volunteers and using only a band with 6 FSR sensors.

Customizing hydrothermal properties of inkjet printed sensitive films by functionalization of carbon nanotubes.

Multiwalled carbon nanotubes (MWCNTs) are attractive materials for realizing sensors, owing to their high aspect ratio associated with excellent mechanical, electronic, and thermal properties. Moreover, their sensing properties can be tuned by introducing functional groups on their framework and adjusting the processing conditions. In this paper, we investigate the potential of functionalized CNTs for humidity and temperature sensing by optimization of the functionalization, the processing conditions and the printing conditions. The morphology of the differently functionalized MWCNTs is investigated by infrared spectroscopy (IR), scanning electron microscopy, thermogravimetry (TG) and TG-coupled mass-spectrometric studies. Using the functionalized MWCNTs, films were fabricated with different numbers of layers (4, 6, 8, 10 layers) via inkjet printing on a flexible polyimide substrate containing an interdigital microelectrode. The influence of hydrothermal effects was investigated. The sensitivity to humidity is higher for films prepared with MWCNTs functionalized with a high sonication amplitude and a bigger number of layers due to enhancements of hydrophilicity and water mobility. A higher sensitivity to temperature is achieved by a low sonication amplitude and a small number of layers. For the encapsulation of the temperature sensor against humidity, a Bectron layer is proposed, which reduces also the hysteresis effect. This study demonstrates the efficiency of carboxylic functionalized MWCNTs deposit by inkjet printing for realization of sensitive and cost-effective humidity and temperature sensors. It provides a real example for the interesting contribution of functionalization procedures to the sensing properties of MWCNTs films.