An Efficient Method for Electric Meter Readings Automatic Location and Recognition

Abstract An efficient and robust meter readings location and recognition algorithm plays a very important role in remote meter reading systems and automatic meter inspection systems. A novel and efficient electric meter readings automatic location and recognition method is presented in this study. It includes the following three steps: coarse-to-fine detection of meter readings, projection-based character segmentation and BP neural network recognition. The algorithm was tested on HUALI, KEYUAN, JINLING, LANJIER and SANXING electric meter images. Experimental results show that the proposed method has satisfying performance and good robustness.

[1]  Yuan Li,et al.  Meter character recognition method based on gray template matching , 2010, Proceedings of the 29th Chinese Control Conference.

[2]  Yoshio Hirose,et al.  Backpropagation algorithm which varies the number of hidden units , 1989, International 1989 Joint Conference on Neural Networks.

[3]  Vera Kurková,et al.  Kolmogorov's theorem and multilayer neural networks , 1992, Neural Networks.

[4]  Eric Lecolinet,et al.  A Survey of Methods and Strategies in Character Segmentation , 1996, IEEE Trans. Pattern Anal. Mach. Intell..

[5]  Shutao Zhao,et al.  Research on Remote Meter Automatic Reading Based on Computer Vision , 2005, 2005 IEEE/PES Transmission & Distribution Conference & Exposition: Asia and Pacific.

[6]  Laurene V. Fausett,et al.  Fundamentals Of Neural Networks , 1994 .