Multi-level neural network based characters identification method and system

Abstract

A system and method, which enable precise and automatic identification of characters, perform and calibrate data verification to ensure data reliability. The system can process these identified characters, such as override adverse conditions, adjusting and correcting unclear characters and their images.

Description

FIELD OF THE INVENTION [0001] The present invention relates generally to a method and system for providing automatic characters recognition. More specifically, the present invention relates to a multi-level neural network based characters identification method and system. BACKGROUND OF THE INVENTION [0002] Systems based on character recognition are widely used in both home and office functions. For example scanners, which reads printed text and translate the image being scanned to a form of data, the computer is able to process (such as ASCII code, for example). Character recognition systems include an optical camera for photographing images, and sophisticated software for analyzing images. The arising need for identifying and analyzing text, characters or other signifying marks on a variety of products, vehicles and aircrafts led to the character recognition system expansion and the system being put to use in different areas of interest. [0003] An operative system at the entrance t...

AI Technical Summary

Benefits of technology

[0014] Therefore it is an object of the present invention to provide fast and reliable methods to enable read and decipher characters on site. It is a further object of the present invention to provide methods that can override adverse conditions and be able to adjust and correct partially unclear images.

[0015] It is a further object of the present invention to provide methods that can perform and calibrate data verification to ensure data reliability.

[0027] The character identification method as described in the present invention combines several ways of identification creating a uniform method providing fast and accurate identification as well as relevant answers to several diverse problems in the identification process that has not been regarded or solved in any of the original methods it is based upon. Thus the uniform system created provides a fitting answer to the needs of today's security and identification systems which demands an operation system that is both fast and accurate. The identification system, despite its complexity and flexibility, is based on relatively simple functions and variables enabling the system to operate on ordinary, not overly complex, computing systems.

Problems solved by technology

The main drawback of this method, and several others in the field, which limits application considerably, relates to the limited number of identifying marks recognized and processed by the method.

Additionally, none of the systems in use today include a method for verification and control of the information retrieved to assess whether the identification is indeed accurate.

This lack of inspection and verification greatly question the credibility of these inadequate systems.

The process is expensive and time consuming and does not always result in a successful and reliable identification.

Another major drawback when considering these systems is the need for a strong computer to run the identification programs.

This is largely due to the complex algorithm employed, which is based on complicated functions and variables, which take long to run even on the strongest of computing systems.

Images