Multiwavelet Framework for Automatic Energy Meter Reading Recognition
In this article, multiwavelet has been utilized for the energy meter reading application. The framework of the proposed algorithm is based upon the multiwavelet and empirical mode decomposition (EMD). Multiwavelet decomposed the image and EMD is used to find the wave crest from the wavelet projected information which provide the horizontal and vertical detail of the energy meter display area from the captured images. A feed-forward neural network is used to recognize the segmented digits of the extracted energy meter display. This work can be further utilized to provide automatic billing purpose. Proposed strategy has been tested on various company energy meters and in variety of situations such as shadow, blurry, dirty etc. The proposed system recognizes meter reading with satisfactory accuracy in less time.
Keywords: Multiwavelet, Projection, Image Enhancement, Energy Meter.
Keywords - Multiwavelet, Projection, Image Enhancement, Meter Reading Recognition.