Water meter recognition method based on dynamic image processing

Abstract

The invention belongs to verification of a water meter, and particularly relates to a water meter recognition method based on dynamic image processing. The method comprises the steps of installing a camera, acquiring a picture, preprocessing, graying and binarizing, positioning a pointer and calculating an angle, acquiring water consumption and the like. The method solves the problem that the recognition result is inaccurate in the prior art, and has the advantages that data has relatively high accuracy and the like in the water meter verification process.

Description

technical field [0001] The invention belongs to the verification of water meters, in particular to a water meter identification method based on dynamic image processing. Background technique [0002] The water meter is a measuring instrument under the national key management. It is widely used in the measurement of water for urban residents and industrial production. It is a measuring instrument for trade settlement. Obviously, the accuracy of water meter measurement is very important, and water meter verification is a necessary means to ensure the accuracy of water meter measurement. . The verification of pointer water meters is based on the "JJG162-2009 Verification Regulations for Cold Water Meters" promulgated by the General Administration of Quality Supervision, Inspection and Quarantine of the People's Republic of China on April 30, 2009. There are three main categories, namely, appearance and function inspection, sealing inspection, and indication error verification....

AI Technical Summary

Problems solved by technology

However, this solution also has shortcomings: First, the quality of the captured photos is easily disturbed by the environment, such as the brightness of the light will affect the reading effect of the photos, and the water droplets in the dial may even cause the photos taken to be unusable for computer recognition. The second is that only static dials can be photographed, and the verification method is limited to the start-stop method, which is not easy to improve the verification efficiency; the third is that the recognition accuracy is low, sometimes individual pointer circles in the dial are blocked, and the computer cannot recognize the data on the dial at all , the versatility is not high

In this method, first record and store the relative angle value a of each picture taken i (The relative angle value is the angle value of the pointer relative to the reference zero degree), and then compare the rotated angle values ​​of two consecutive pictures in turn, and store all the compared rotated angle values ​​in turn, that is, the absolute angle value b i , and each calculation of the absolute angle value of two consecutive pictures will bring a certain error, and finally the absolute angle value b i Add up to calculate the total number of turns m that the pointer has turned. Multiple error accumulation transfers will increase the calculated error value and reduce the accuracy of the water consumption identification of the water meter.

Images