Overview
ISO 1073-2:1976 specifies the shapes and dimensions of the printed image for the OCR‑B alphanumeric character set used in optical character recognition (OCR). Published by ISO in 1976 (corrected/reprinted 1979), the standard defines a machine- and human‑readable repertoire aligned with ISO 646‑1973 (7‑bit coded character set) and establishes a common basis for printing and optical scanning equipment to support OCR interchange.
Key topics and technical requirements
- Character repertoire: OCR‑B comprises 121 characters with defined sub‑sets (numeric, initial alphanumeric, extended alphanumeric, options and erase characters). The numeric sub‑set contains 22 characters; other sub‑sets allow progressively larger character coverage.
- Fonts and styles: Two styles are defined:
- Constant‑strokewidth font (definitive centrelines and constant stroke width).
- Letterpress font (variable stroke endings for finer printers; centrelines remain the same).
- Sizes and scaling: OCR‑B is specified in three sizes (designated sizes I, III and IV) with scale factors relating sizes III and IV to size I (vertical and horizontal factors are given). Centreline heights for the character “EIGHT” are specified (e.g., size I = 2.40 mm, size III = 3.20 mm, size IV = 3.60 mm).
- Reference grid & centrelines: Character shapes and dimensions are defined by reference drawings on a grid. The centreline positions are the definitive element for OCR interpretation.
- Strokewidth and tolerances: Nominal strokewidth is constant per character for the constant‑strokewidth font; some printing processes may limit corner sharpness-corner radii need not be smaller than 0.08 mm for OCR reliability.
- Erase characters: Two erase symbols (character erase and group erase) are defined with minimum dimensions for reliable recognition.
- Interchange guidance: The standard addresses cross‑border document circulation and advises agreement where different national characters are used.
Applications and who uses it
ISO 1073‑2:1976 is intended for:
- OCR system designers and software developers (character recognition rules, training sets).
- Printing equipment manufacturers and typefoundries (character shapes, stroking, and font production).
- Operators of devices such as typewriters, cash registers, high‑speed printers, numbering machines, credit‑card imprinters, and offset/letterpress printers.
- Organizations implementing automated data capture in banking, postal, transportation and administrative document workflows that require standardized, machine‑readable text.
Related standards
- ISO 1073‑1 (OCR‑A: numeric and character dimensions) - Part I of the OCR family.
- ISO 646‑1973 - ISO 7‑bit coded character set referenced for character repertoire.
Open preview (PDF)Alphanumeric character sets for optical recognition — Part 2: Character set OCR-B — Shapes and dimensions of the printed image