Gesture recognition, as a core technology of human–computer interaction, has broad application prospects and brings new technical possibilities for smart homes, medical care, sports training, and other fields. Compared with the traditional human–computer interaction models based on PC use with keyboards and mice, gesture recognition-based human–computer interaction modes can transmit information more naturally, flexibly, and intuitively, which has become a research hotspot in the field of human–computer interaction in recent years. This paper described the current status of gesture recognition technology, summarized the principles and development history of electromagnetic wave sensor recognition, stress sensor recognition, electromyographic sensor recognition, and visual sensor recognition, and summarized the improvement of this technology by researchers in recent years through the direction of sensor structure, selection of characteristic signals, the algorithm of signal processing, etc. By sorting out and comparing the typical cases of the four implementations, the advantages and disadvantages of each implementation and the application scenarios were discussed from the two aspects of dataset size and accuracy. Based on the abovementioned discussion, the problems and challenges of current gesture recognition technology were discussed in terms of the biocompatibility of sensor structures, wearability and adaptability, stability, robustness, and crossover of signal acquisition and analysis algorithms, and the future development directions in this field were proposed.

Flexible wearable sensors, characterized by their stretchability, flexibility, lightweight, high sensitivity, and the ability to dynamically and continuously measure multiple parameters, have been widely applied in areas such as physiological health monitoring, human-computer interaction control, real-time motion monitoring, and personalized medical assistance. With the development of artificial intelligence technology, the integration of flexible wearable sensors with artificial intelligence technologies such as machine learning has become increasingly profound. This integration has facilitated the emergence and development of intelligent sensing systems with human-like perception capabilities, garnering growing attention towards flexible wearable intelligent sensing technology. This paper focuses on introducing various sensor mechanisms, intelligent sensing algorithms, and practical applications of flexible wearable intelligent sensing technology. The challenges and possible development directions of flexible wearable smart sensing technology, encompassing aspects such as power supply, multimodal sensing capability, preparation materials, processing and fabrication techniques, and smart sensing algorithms, are also discussed and prospected. Such a review can provide a reference for research in the field of flexible wearable smart sensing technology.

Sapphire fiber, renowned for their high melting point, mechanical robustness, and superior optical properties, are extensively utilized in high-temperature sensing applications. Sapphire fiber Bragg gratings (SFBG) fabricated from these fibers exhibit a high refractive index, a large mode field diameter, and a lack of cladding, which collectively result in significant multimode propagation. This multimodal behavior has the effect of broadening the reflection peak bandwidth, which in turn adversely impacts the precision of the measurements made by the sensor. To solve this problem, the present study employs a double-conical splicing technique to achieve quasi-single-mode SFBG, effectively suppressing the multimode effect and successfully realizing real-time demodulation up to 1,700 °C. Meanwhile, to address the impact of vibrational crosstalk and background radiation on measurement accuracy in high-temperature environments, this study introduces a PCA-KNN algorithm for demodulation, significantly improving demodulation accuracy in the high-temperature range. Experimental results demonstrate that the sensors exhibit high sensitivity (37 pm/°C) at elevated temperature, a high measurement limit (1,700 °C), and improved demodulation accuracy (2.42 °C), while successfully suppressing vibrational crosstalk and radiation interference. This research provides a novel solution for high-precision real-time monitoring and the practical application of SFBG sensors.

Hand posture detection in underwater environments is a crucial method for human-computer interaction, widely applied in areas such as diving rescue, underwater engineering, and marine scientific exploration. Vision-based gesture recognition systems are susceptible to ambient light and water turbidity, while other electrical sensing technologies often lack sufficient sensitivity and reliability under dynamic underwater conditions. To address these challenges, we designed a wearable data glove based on Fiber Bragg Grating (FBG) technology, called the FBG-SenseGlove. We utilized food-grade silicone to achieve integrated encapsulation of the FBG sensors within the glove. In strain tests, the sensors exhibited a sensitivity of 1.2 pm/με. In bending tests of the glove, the linearity was 0.98961, the bending sensitivity was 1.8796 pm/°, and the maximum wavelength shift after bending reached 1.7 nm, demonstrating excellent stability and sensitivity. We also designed 16 common underwater communication gestures and six different complex usage scenarios, collecting datasets using the FBG-SenseGlove. A self-designed GRU-FusionNet deep learning network model was employed to extract, fuse, and classify the temporal features of the dataset. The results show that the validation set accuracy for gesture recognition in undisturbed normal scenarios was 99.48%, and the accuracy on the test sets in complex scenarios reached 98.98% after model transfer. Our FBG-SenseGlove, combined with deep learning algorithms, operates stably even in complex environments, providing theoretical validation and technical support for underwater gesture communication and specialized tasks in extreme conditions.

Flexible pressure sensors, as a key branch of flexible electronics, have garnered widespread attention in various fields due to their exceptional compatibility and flexibility. A large number of studies have been conducted to design flexible pressure sensors with excellent performance by utilizing different materials, structures, and emerging manufacturing processes. This paper outlines the performance metrics used to evaluate flexible pressure sensors, as well as various sensing mechanisms, and provides a detailed introduction to emerging materials and technologies for manufacturing. It also focuses on the latest research on the production of flexible pressure sensors using microelectronic printing techniques, including screen printing, inkjet printing, and dispensing printing. In addition, this work provides a comprehensive overview of recent advances in the applications of flexible pressure sensors across various fields, including healthcare, sports, e-skin, human–machine interaction, underwater environments, and intelligent sensing algorithm–based systems. Finally, the paper discusses the challenges and future development prospects, aiming to provide comprehensive, timely, and scientific references for the field of flexible pressure sensors.

Fiber optic haptic skin is a class of wearable devices that mimic the structure of human skin tactile sensing system, which is gradually replacing electronic skin as a new application solution in health monitoring and human-computer interaction due to its advantages of high sensitivity and resistance to electromagnetic interference. However, most of the current fiber-optic haptic skins use a single optical fiber as a point sensing unit, which cannot accurately sense the vectorial angle information of the stress when touching complex textures. Here, we explore an enhanced structure for texture angle information extraction based on a multi-axial haptic sensing unit with dual micro-nano optical fibers, which can better differentiate between normal and axial forces, greatly expanding its potential use in the direction of haptic sensing. First, we experimentally verified the effect of the waist region crossing angle α on the vector information extraction performance of the anisotropic sensing unit, and verified that α equal to 90° is the optimal solution for the sensor structure. Further, we discuss the ability of traditional mathematical statistics and machine learning methods to construct the vector decoupling reconstruction model of force and optical power loss in this experiment through simulation and experiment. The results show that the machine learning method constructs better, and the average error is about 10% of the error value of the multiple linear regression algorithm, which is only 0.0347 dB. As a validation, we combined the machine learning algorithm to test its ability to categorize the vein textures on different leaf surfaces with an accuracy of 98.5%. This work lays an important foundation for high-precision fiber-optic texture tactile sensing.