Development of an automatic reading method and software for pointer instruments

Abstract

To effectively resolve the automatic recognition of analog pointer measuring instruments, a reading recognition method based on Hough transform is proposed. In order to simplify the following procedure, image graying and binary, mathematical morphology, including dilation and thinning, edge recognition was applied to the system. We proposed the recognition principle for reading recognition of the pointer's angle by using Hough transform. And the reading of the pointer is decided by the linear relationship between the panel's scale and the angle of the instrument's dial revolutionary. Considering that in most practical cases, the panel cannot be horizontal, so the procedure of tilt correction is employed in this paper. Finally, the software is written in MATLAB. The recognition system is achieved in LabVIEW, where the front panel of the reading system is developed. The experimental results demonstrate that our method is feasible for pointer instrument recognition.