Publication: Automated analog meter reading using angle detection method
Loading...
Date
2021-07-01
Authors
Cheah, Kai Bin
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
In the now-happening 4th industrial revolution and rising of IoT implementation, machine vision is becoming one of the latest technologies that has been gradually used in industries processes to increase the accuracy and speed of the production lines. One of the said processes are the inspection and recording field. Many factories are still using traditional recording method of manually read and write the meter reading on a notebook, and these processes are time consuming, causing wastage of manpower, and tiring. Thus, this work aims to propose and develop an automated analog meter reading system where the picture of number indicated by the analog meter will be captured by camera, analysed, and output with the reading automatically. Method used by the system is through angle detection, where angle between two lines is calculated. The system only needs to find three points to compile an angle. The system can be distributed into two phases, where the first phase involving finding the centre point of meter and the number zero “0” as the reference points. Next, the point at the end of pointer is found and angle is calculated. These three points are chosen as most of the meter on the market has all the characteristics mentioned, making this system more robust. For this thesis, two methods based on different techniques are developed and compared with each other. Both use angle detection to get the actual value, but one uses combination of line and color detection to detect the pointer location, while other uses color detection only. The first method is chosen to be implemented in the system as it shows higher robustness as compared to the second method.