Research On Image Processing And Recognition Algorithm To Direct-Reading Device Of Intelligent Instrument

Contraposing to the various defects of the traditional manual digit meter reading, people bring forward an automatic meter reading technology, and digit character recognition is a very important link, which is the content of this paper.The digit meter image in this paper is shooting by digital cameras, recognition process including: digit image preprocessing, digit character location and character recognition. Image preprocessing including image denoising, image binarization and tilt correction. Against unknown light conditions, The classic binarization algorithms such as Ostu algorithm and Bersen algorithm can not extract object well, so we use S.D.Yanowitz algorithm which interpolates the image grey by gradient information. Then it can get digit meter frame information by horizontal and vertical projection, and extract line by improved Hough transform. Finally, the image can be corrected by affine transform. Digit character location first uses a priori knowledge for approximate location. Then morphologic enhancement and connected component filter make location more precisely. At last, the tangent rectangle can be solved by projection method, and can be differentiated between complete character and half character. Character recognition including entire character recognition and half character recognition two parts. According to digit character characteristic information, the entire character uses the special method of typography digit recognition to identify. The half character recognition uses sliding template matching based on improved Hausdorff distance to identify and both of them obtain a high rate of correct identification. Finally, this paper has confirmed algorithm actual feasible by experiment and gives error analysis which possibly appears.