Liquid Level Detection in Standard Capacity Measures with Machine Vision

Abstract

Capacity measures are commonly used volume standards for testing measuring systems for liquids other than water. Manual readings from the measuring scale can often be difficult due to the location of the capacity measure or to the nature of the measured liquid. This article focuses on the automation of this procedure by using a single camera machine vision system. A camera positioned perpendicular to the transparent neck captures the image of the liquid meniscus and the measuring scale. The volume reading is determined with the user-defined software in the LabVIEW programming environment, which carries out the image preprocessing, detection of the scale marks and the liquid level, correction of lens distortion and parallax effects and final unit conversions. The realized measuring system for liquid level detection in standard capacity measures is tested and validated by comparing the automated measurement results with those taken by the operators. The results confirm the appropriateness of the presented measuring system for the field of legal metrology.