A novel approach of reading analog multimeter based on computer vision

An approach of reading analog multimeter based on computer vision technique is proposed in this paper. This method can detect not only the angle of needle-type pointer but also that of selector switch for identifying the measured value of the selected function from the analog multimeter image. A vision system of reading the linear scale of DCV, ACV, DCmA, and the nonlinear scale of resistance from an analog multimeter were developed for experiments and validating the proposed approach under various scenarios of different illuminations and shifted/tilted situations. Our experiments confirm the feasibility of the proposed approach.