Segmentation-free approaches of computer vision for automatic calibration of digital and analog instruments

A common problem found in the calibration laboratories is the reliability of the results obtained from calibration of instruments, especially when they do not have a built-in communication interface. In this case, the time consuming is increased significantly and the calibration may be subject to human error. Thus, many approaches based on computer vision have been proposed in the literature for automating calibration processes. However, most of them first simplify the images, usually using operations such as segmentation/binarization, for after proceeds the recognition of the digits. The issue lies in the fact that these simplifying operations throw away the rich grayscale information, decreasing the robustness of the algorithms, mainly when the images are affected by illumination changes, noise or JPEG compression. In this paper it was proposed two segmentation-free algorithms for automating the calibration process of digital and analog measuring instruments without built-in communication interface. The first one is based on template matching with normalized cross correlation for reading the display digits and the second uses radial projections and Bresenham algorithm to determine the pointer position in analog instruments. The experimental results showed that the proposed algorithms presented high accuracy and performance and can be used in real time systems applied to calibration processes.

► A common problem in calibration process is the reliability of the obtained results.
► In the last years many methods based on computer vision have been proposed.
► Two segmentation-free algorithms for automating calibration process were proposed.
► The proposed algorithms are robust because they do not simplify the images.
► The proposed algorithms can be used by real time systems for automatic calibration.