Automatic Recognition of Traditional Analog Gauge and Intelligent Management by IOT Method
Chuan-Chun Wu, Hsien-Leing Tsai, Chien‐Hsing Wu, Po-Jen Wang · 2021
The purpose of this research is using the technology of the Internet of Things to improve current medical or traditional industries in meter recording system. This research develops an IoT automatic meter reading system that is practical, precise, and highly feasible. The main device includes Arduino, two-axis slide rail, stepping motor and driver, Raspberry PI4, camera and server PC. Two-axis stepping motor is driven by the Arduino through the motor drive board, so that the camera on the slide rail accurately aligns with the center of the meter surface to improve the deviation cause by reader's viewing angle. After the Arduino positioning is completed, it notifies the Raspberry PI4, and then PI4 issues a screen capture command to the camera. We use OpenCV image recognition software to correctly identify and calculate the best value to achieve a goal that is more accurate than human reading. Finally, the original image and the recognized value are sent back and save on the PC server database (My SQL) through the TCP/IP Socket which is running on the Raspberry PI4. And then, Big Data analysis is applied, and its result is presented via Web-Service Page. This saves manpower trip or avoid the risk of entering a dangerous area and realize unmanned gauge recording. And when the system equipment is upgraded, there is no need to replace the original equipment components to achieve seamless system conversion.