ATM Check MICR Recognition Using MR Waveforms and OCR

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing automated teller machines (ATMs) face challenges in accurately recognizing magnetic ink characters on checks without requiring a high-resolution MICR sensor, leading to potential counterfeit detection failures and increased manufacturing costs.

Innovation Solution

A method utilizing both an MR sensor and a contact image sensor to compare optical character recognition (OCR) results with character-based waveforms, employing a waveform authentication algorithm to determine the correct grouping of magnetic ink characters, thereby eliminating the need for a high-resolution MICR sensor.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If a high-resolution MICR sensor is provided in the authentication unit to correctly read magnetic ink characters on checks, then the recognition accuracy of magnetic ink characters is improved, but the device complexity and manufacturing cost increase

Engineering Contradiction:
Improverecognition accuracy of magnetic ink charactersVSAvoidauthentication unit configuration
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent combines the functions of magnetic ink character recognition and bill authentication into a single authentication unit. The MR sensor originally designed for bill authentication is merged with OCR processing capabilities to handle both check and bill authentication, eliminating the need for a separate high-resolution MICR sensor

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The authentication unit is designed with multi-functionality to handle both bill authentication and check authentication. The same authentication unit uses the MR sensor for magnetic signal detection and incorporates OCR algorithms to recognize magnetic ink characters on checks, as well as authenticate bills, making the system versatile without requiring additional specialized sensors

Inventive Principle:
Principle #6Universality (Multi-functionality)

2Measurement precision

If a high-resolution MICR sensor is provided in the authentication unit to correctly read magnetic ink characters on checks, then the recognition accuracy of magnetic ink characters is improved, but the manufacturing cost increases

Engineering Contradiction:
Improverecognition accuracy of magnetic ink charactersVSAvoidmanufacturing cost
Core Design Contradiction:
Measurement precisionVSEase of manufacture

Solution Approach 1:

The authentication unit is designed with multi-functionality to handle both bill authentication and check authentication. The same authentication unit uses the MR sensor for magnetic signal detection and incorporates OCR algorithms to recognize magnetic ink characters on checks, as well as authenticate bills, making the system versatile without requiring additional specialized sensors

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The system uses its existing MR sensor and processing capabilities to perform magnetic ink character recognition without requiring external specialized equipment. The authentication unit leverages its own magnetic sensing capabilities and computational resources to execute OCR algorithms, making the system self-sufficient

Inventive Principle:
Principle #25Self-service

3Reliability

If results of OCR and character-based waveforms are compared for verification to recognize magnetic ink characters, then the reliability of check authentication is improved, but the processing time increases

Engineering Contradiction:
Improvereliability of check authenticationVSAvoidprocessing time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The system implements feedback by comparing OCR results with character-based waveforms from the MR sensor. This verification process uses the magnetic waveform data as feedback to validate or correct the OCR recognition results, improving reliability through cross-validation of the two recognition methods

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The verification process applies partial action by selectively comparing only critical characters or positions where errors are most likely to occur. The system performs waveform comparison verification on key magnetic ink characters rather than every character, balancing reliability improvement with processing time constraints

Inventive Principle:
Principle #16Partial or excessive action

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

Ensures accurate recognition of magnetic ink characters, maintaining authentication reliability while reducing the complexity and cost of the ATM's authentication unit configuration.

Implementation Method 1

a contact image sensor provided in an authentication unit and character-based waveforms of magnetic ink characters that are acquired through an MR sensor

Methodology Applied
Scientific EffectPhotoelectric Effect: Photoelectric Effect

Implementation Method 2

character-based waveforms of magnetic ink characters that are acquired through an MR sensor provided in order to authenticate whether a bill is legitimate

Methodology Applied
Scientific EffectMagnetic Resonance:

Data Source

PatentEP4276780B1Method of recognizing magnetic ink characters in ATM
Publication Date: 2026.04.29 HYOSUNG TNS INC
  • EP4276780B1 patent drawingFigure 1~2
  • EP4276780B1 patent drawingFigure 3~4

AI summary

The present disclosure relates to a method of recognizing magnetic ink characters in an automated teller machine and, more particularly, to a method of recognizing magnetic ink characters in an automated teller machine, the method being capable of comparing, for verification, results of optical character recognition (OCR) on the basis of an image acquired through a contact image sensor (CIS) provided in an authentication unit and character-based waveforms of magnetic ink characters that are acquired through an MR sensor provided in order to authenticate whether a bill is legitimate and thus of realizing correct reading of magnetic ink characters printed on a check while an automated teller machine handling both a bill and a check reads the magnetic ink characters printed on the check inserted thereinto. With this method, a structure of the authentication unit can be simplified in such a manner that correct recognition of the magnetic ink characters printed on the check in a normal automated teller machine is supported without a high-resolution MICR sensor required to be provided in the normal authentication unit in order to read the magnetic ink characters printed on the check. Moreover, the cost of manufacturing the automated teller machine can be effectively reduced.