At present,mechanical water meter verification devices still rely on manual operation.There are many prob-lems in the verification process,such as long time consuming,high labor cost and low efficiency.In order to improve the efficiency of mechanical water meter verification,a reading recognition system based on machine vision is designed in this paper.The proposed system obtains the picture of the mechanical water meter by reading the real-time video stream,and then adopts the convolutional neural network to build the reading recognition system of the water meter.The proposed sys-tem identifies the reading of the mechanical water meter according to the water meter picture obtained by the video stream.Finally,the effectiveness of the proposed system is verified by experiments.
关键词
机械式水表/卷积神经网络/机器视觉
Key words
mechanical water meter/convolutional neural network/machine vision
维普
万方数据