US 7,480,403 B2Granted

Apparatus, system, and method for fraud detection using multiple scan technologies

AssigneeInternational Business Machines Corporation

InventorsRobert William Kruppa, Ravinder Prakash

Life of this patent

Estimating the term…

Loading patent…

Abstract

An apparatus, system, and method are disclosed for detecting fraud. A first optical character recognition module optically recognizes a character imprinted with a magnetic ink on a medium using a first optical character recognition algorithm. A second optical character recognition module also optically recognizes the character using a second optical character recognition algorithm. In addition, a magnetic ink character recognition module magnetically recognizes the character using a magnetic recognition algorithm. A voting module determines if the character is potentially fraudulent based on the recognition results of the first optical character recognition module, the second optical character recognition module, and the magnetic ink character recognition module. If the character is potentially fraudulent, a results module communicates a fraud indicator.

Claims27

Dependent claims

An apparatus to detect fraud, the apparatus comprising:

an MICR module that scans a single row of Magnetic Ink Character Recognition (“MICR”) characters imprinted on a medium using magnetic ink, recognizes individual MICR characters within the single row of MICR characters, and detects a magnetic signal strength of the magnetic ink of the MICR characters;
a first OCR module that scans the single row of MICR characters and recognizes individual MICR characters within the single row of MICR characters, the first OCR module recognizes the individual MICR characters using a first Optical Character Recognition (“OCR”) algorithm;
a second OCR module that scans the single row of MICR characters and recognizes individual MICR characters within the single row of MICR characters, the second OCR module recognizes the individual MICR characters using a second OCR algorithm, wherein the second OCR algorithm differs from the first OCR algorithm;
a voting module that determines if a MICR character is potentially fraudulent based on the recognition results of the MICR module, the first OCR module, and the second OCR module, wherein determining that a MICR character is potentially fraudulent comprises one or more of the MICR module determining that the magnetic signal strength of the MICR character is lower than a predetermined value and the character recognized by the MICR module, the character recognized by the first OCR module, and the character recognized by the second OCR module are not identical;
and
a results module configured to communicate a fraud indicator if the voting module determines that a MICR character is potentially fraudulent,
wherein the MICR module, the first OCR module, the second OCR module, the voting module, and the results module comprise one or more of logic hardware and a processor executing executable code, the executable code stored on one or more computer readable media.

4 more elements

An apparatus to detect fraud, the apparatus comprising:

a communication interface module configured to receive one or more optical scans and a magnetic scan of a row of characters imprinted on a check using magnetic ink;
an MICR module that scans a single row of MICR characters imprinted on a medium using magnetic ink, recognizes individual MICR characters within the single row of MICR characters, and detects a magnetic signal strength of the magnetic ink of the MICR characters;
a first OCR module that scans the single row of MICR characters and recognizes individual MICR characters within the single row of MICR characters, the first OCR module recognizes the individual MICR characters using a first OCR algorithm;
a second OCR module that scans the single row of MICR characters and recognizes individual MICR characters within the single row of MICR characters, the second OCR module recognizes the individual MICR characters using a second OCR algorithm, wherein the second OCR algorithm differs from the first OCR algorithm;
a voting module that determines if a MICR character is potentially fraudulent based on the recognition results of the MICR module, the first OCR module, and the second OCR module, wherein determining that a MICR character is potentially fraudulent comprises one or more of the MICR module determining that the magnetic signal strength of the MICR character is lower than a predetermined value and the character recognized by the MICR module, the character recognized by the first OCR module, and the character recognized by the second OCR module are not identical;
and
a results module configured to communicate a fraud indicator if the voting module determines that a MICR character is potentially fraudulent,
wherein the MICR module, the first OCR module, the second OCR module, the voting module, and the results module comprise one or more of logic hardware and a processor executing executable code, the executable code stored on one or more computer readable media.

5 more elements

A system to detect fraud, the system comprising:

a memory module;
a processor module;
and
a detection module comprising:
an MICR module that scans a single row of MICR characters imprinted on a check using magnetic ink, recognizes individual MICR characters within the single row of MICR characters, and detects a magnetic signal strength of the magnetic ink of the MICR characters;
a first OCR module that scans the single row of MICR characters and recognizes individual MICR characters within the single row of MICR characters, the first OCR module recognizes the individual MICR characters using a first OCR algorithm;
a second OCR module that scans the single row of MICR characters and recognizes individual MICR characters within the single row of MICR characters, the second OCR module recognizes the individual MICR characters using a second OCR algorithm, wherein the second OCR algorithm differs from the first OCR algorithm;
a voting module that determines if a MICR character is potentially fraudulent based on the recognition results of the MICR module, the first OCR module, and the second OCR module, wherein determining that a MICR character is potentially fraudulent comprises one or more of the MICR module determining that the magnetic signal strength of the MICR character is lower than a predetermined value and the character recognized by the MICR module, the character recognized by the first OCR module, and the character recognized by the second OCR module are not identical;
and
a results module configured to communicate a fraud indicator if the voting module determines that a MICR character is potentially fraudulent.

7 more elements

A computer readable medium tangibly embodying a program of machine-readable instructions executable by a digital processing apparatus to perform operations to detect fraud, the operations comprising:

magnetically scanning a single row of Magnetic Ink Character Recognition (“MICR”) characters imprinted on a medium using magnetic ink;
recognizing individual MICR characters within the single row of MICR characters and detecting a magnetic signal strength of the magnetic ink of the MICR characters, wherein recognizing MICR characters and detecting the magnetic signal strength comprise using a magnetic recognition algorithm;
optically scanning the single row of MICR characters
recognizing individual MICR characters within the single row of MICR characters using a first OCR algorithm;
recognizing individual MICR characters within the single row of MICR characters using a second OCR algorithm, wherein the second OCR algorithm differs from the first OCR algorithm;
determining if a MICR character is potentially fraudulent based on the recognition results of the first OCR algorithm, the second OCR algorithm, and the magnetic recognition algorithm, wherein determining that a MICR character is potentially fraudulent comprises one or more of determining that the magnetic signal strength of the MICR character is lower than a predetermined value and the character recognized by the magnetic recognition algorithm, the character recognized by the first OCR algorithm, and the character recognized by the second OCR algorithm are not identical;
and
communicating a fraud indicator in response to determining that a MICR character is potentially fraudulent.

5 more elements

Dependencies are inferred from the claim text.