A deep-learning-assisted versatile electret sensor for moving object detection

Smart active sensing system plays an important role in the development of smart industry and Internet of Things (IoT). For large-scale machinery and equipment, it's still very challenging to develop ideal sensors for accommodating to this future technology. In this work, a self-powered versatile electret sensor is proposed for moving objects detection in large-scale machinery and equipment, based on which, methods of detecting the features of objects, such as size, material composition, charged property, length and speed, are investigated and developed. Deep learning method is also integrated with the object sensor, reaching very high accuracy and intelligent sensing. Moreover, take the case of aircraft inlet or other machinery with similar issues, we further design a specific object sensor for inhalation monitoring. An intelligent monitoring and recognition system based on object sensor and deep learning method is also designed for intelligent and automatic monitoring of moving objects. Very high accuracy is obtained, indicating promising potential of this prototype in smart sensing. This work provides a deep-learning-assisted self-powered versatile electret sensor for moving objects, which lays a solid foundation for the real-time health and safety monitoring and digitalization of large-scale machinery and equipment in IoT.

Automated summary

This research proposes a novel self-powered versatile electret sensor for detecting moving objects in large-scale machinery. By integrating deep learning methods, the sensor achieves high accuracy and intelligent sensing. Furthermore, a specific object sensor is designed for inhalation monitoring. The study provides crucial support for real-time health and safety monitoring and digitalization of large-scale machinery.