A YOLOv3-Based Industrial Instrument Classification and Reading Recognition Method

Aiming at the demand of industrial instrument reading, this study proposes a method classification and reading recognition based on YOLOv3. Given that meters can be divided into pointer digital according to dial type, conducts for each two types meters. Firstly, YOLOv3 model is trained recognize detect meter classify values obtained classes. The uses Hough circle dial, extracts scale pointer, calculates angle between 0 line obtains meter. digits by finding contours digit area then support vector machine (SVM) identify extracted output readings Through test, mean average precision (mAP) in 93.73%. absolute error less than 0.1 general, maximum relative 0.35%. accuracy 99.7%. proposed accurately read value meet needs production.