Optimized meter calibration method based on auto recognition

Periodical calibration is the important method for making instruments in good condition. In this paper, a new measurement instruments calibration method is introduced, which adapts a novel scheme following the techniques of the computer vision discrimination, incorporate human´s ability of visual and apprehensive models, various meter calibration can be optimized automatically. The central idea of the proposed method is to capture the measuring instrument images in realtime when the standard source outputting analog, the location of needle and dial can be ascertained using computer image recognition, and then the computer control the analog to the next instrument scale value automatically, up to the full-scale. In the meter calibration process, the needle localization can be successfully retrieved in every output analog value with complex needle movement of the whole course, and calibration results are given automatically according to the measure errors.