Numeral Recognition Of Power Meter On A Handheld Terminal

Le résumé fourni par la source

Abstract- In this paper,we present a new method of numeral recognition that applies character features of the power meter images taken by a handheld terminal camera.The main difficulties for this recognition arise from pictures of low acquisition resolution,noise,heavy or light priting,and numerial distortion.Besides,different dials have different figures in sizes and font.Based on above difficulties, a new kind of 8 Direction Tracking Find-Hole method is proposed. Numbers are recognized by extracting holes and gaps of each character. The systerm gets binary image firstly by means of preprocessing and then extracts holes and gaps of target figures based on structural characteristics.The Find-Hole method in this paper dosen’t have to normalize irregular characters.Experiments show that this method performs well in anti-interference, reducing the complexity of the algorithm effectively and enhancing the processing speed and recognition rate. applying the algorithm above.Section Ⅳ disscusses results of the application and,finally in section Ⅴ we draw the conclusions from this work. Index Terms—Image Processing;numerical recognition; holes;gaps; architectural feature I.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.