Design of Contactless Intelligent Epidemic Prevention System based on PYNQ

ABSTRACT

To combat the COVID-19 virus, extensive studies are being conducted on contactless alternative methods for measuring body temperature and detecting whether masks are being worn. This study investigated a system that employed ZYNQ-7020 as the main controller. The accurate non-contact measurement of a persons body temperature was realized by the calibration and compensation of infrared sensor data collected by an MLX90614 thermopile. In addition, OpenCV and a convolutional neural network (CNN) algorithm were used for face image recognition and to detect whether a person was wearing a mask. A mask detection algorithm was developed based on high-level synthesis and implemented by Python productivity for Zynq (PYNQ). The PYNQ test results were output through a high-definition multimedia interface. Users could also view the real-time human body temperature and face image recognition results through the web. The system achieved a temperature measurement distance of 30 cm, with a measurement error of ± 0.3 ◦C and measurement range of 2560 ◦C. The system also had an alarm function with a buzzer that could be used for an over-temperature alert. The algorithm used the characteristics of field programmable gate array parallel computing to improve the CNN calculation in image processing. A variety of optimization strategies were implemented to achieve hardware acceleration, which improved the mask recognition rate. [ FROM AUTHOR] Copyright of Engineering Letters is the property of Newswood Limited and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full . (Copyright applies to all s.)