MICR Reader Software for Check Processing

A MICR (Magnetic Ink Character Recognition) reader is a device or software that reads the special magnetic ink characters printed at the bottom of checks. These characters contain the routing number, account number, and check number. MICR readers use either magnetic sensors to detect the magnetic properties of the ink, or optical recognition combined with machine learning to read the characters visually. Modern check scanning solutions combine both approaches for maximum accuracy.

MICR reading specifically targets the magnetic ink characters at the bottom of checks (E-13B or CMC-7 fonts), while OCR (Optical Character Recognition) reads any printed text visually. MICR was designed for high-speed, high-accuracy reading of bank documents even with overprinting or stamps. MICR provides 99.9%+ accuracy for the MICR line, whereas OCR is used for other fields like payee names and amounts. Modern systems use both technologies together.

The MICR line at the bottom of a check contains three key pieces of information: (1) Routing Number (9 digits) identifies the bank and branch where the account is held; (2) Account Number identifies the account holder's account; (3) Check Number identifies this specific item. In North America, this data is formatted using the E-13B font with special delimiter symbols separating the fields.

E-13B is the MICR font standard used in North America, Australia, and the UK. It uses 14 characters (0-9 and 4 special symbols) with a specific magnetic waveform pattern. CMC-7 is used in most European, Latin American, and some Asian countries. It features a 7-bit barcode-style character set with more distinct visual patterns. Chequedb's check processing API automatically detects and reads both font types.

Yes. Modern MICR readers use advanced optical recognition combined with AI to read MICR characters even from poor quality images, mobile photos, faded ink, or partially obscured lines. While traditional magnetic readers require clean magnetic signals, optical MICR reading can handle degraded images by using machine learning models trained on millions of check samples. Chequedb's system also includes image preprocessing to enhance quality before reading.

Automated MICR reading achieves 99.9%+ accuracy for the MICR line when using magnetic reading, and 99.5%+ with optical recognition. The technology is specifically designed for high-reliability financial processing. Errors are extremely rare because MICR fonts have built-in error detection — each character produces a unique magnetic waveform that's difficult to misread. Chequedb validates extracted data against routing number databases for additional verification.