Image recognition based automatic reading method for multi-pointer meter
Abstract：Aiming at the recognition of pointer reading for multi-pointer meter,a positioning method based on chromatism transformation as well as an automatic recognizing and reading method based on structural characteristics of a dial were proposed.The luminance and red color difference（LRCD） transform,mathematical morphology based processing method and p-tile method were used to accomplish the selection of effective area and the pointer positioning of water meter.According to such features as the angle between the linking line of dial center to pointer rotatory center and the zero graduation mark of sub-dial,the angle between the zero graduation mark of sub-dial and the pointer directed line was calculated,and furthermore,the pointer reading was recognized and read.The experimental results show that the proposed algorithm for positioning and recognizing has calculating simpleness and higher accuracy,and can avoid the effect of random angle tilt of dial on reading recognition algorithms.
Keywords：automatic meter readingpointer readingimage recognitionLRCD transformmathematical morphologypointer positioningtilt correctionrotatory center
Publication Date：2011-01-01
Online Publishing Date：2025-08-15(First online date of this platform, not the publication date of the document)