Barcode symbologies

Single dimension codes (linear)

Numeric codes

UPC: Universal Product Code; UPC A, UPC E; used by North American Distribution Industry
EAN: European Article Numbering system; EAN 8, EAN 13
JAN: Japanese Article Numbering system; JAN 8, JAN 13; Character set: 0-9

Interleaved 2 out of 5: aka Material Handling Institute standard USD-1
USD-1: Uniform Symbol Description 1; Character set: 0-9

Alphanumeric

Codabar

Two versions:

MHI standard USD-4

American Blood Commission (ABC) standard

Character set: $+-./0123456789: and ABCD (used for Start/Stop codes only)

3 of 9 code: aka MHI standard USD-3; aka Code 39; Character set: (SPACE) $%+-./ and 0-9 and A-Z and “*” (used for start/stop codes only)

Extended 3 of 9 code: provides full 128 character ASCII set; aka Extended code 39; Characters $/+% followed by a upper-case letter can represent other ASCII characters; ASCII 000 (NULL): %U; ASCII 001-026: $A-$Z; ASCII 027-031: %A-%E; ASCII 032 (space): (space); ASCII 033–044 !”#$%&’()*+, : /A-/L; ASCII 045 -: - (also /M ??); ASCII 046 .: . (also /N ??); ASCII 047 /: /O; ASCII 048-57 (0-9): 0-9 (also /P-/Y); ASCII 058 :: /Z; ASCII 059 ;: %F; ASCII 060-064 <=>?@ : %G-%J, %V; ASCII 065-090 A-Z: A-Z; ASCII 091-096 [\]^-` : %K-%O, %W; ASCII 097-122 a-z: +A-+Z; ASCII 123-126 ??}~: %P-%S; ASCII 127 (DEL): %T (also %X, %Y, %Z)

Full ASCII (7-bit)

Code 128

ISO/IEC 15417:2007

108 symbols (103 data, 3 start, 2 stop); 3 black bars and 3 white spaces, each 1–4 modules wide; each symbol is 11 modules wide; sum of bar module widths must be even within each symbol

Overall code must be a minimum of 34 modules wide (54 modules including quiet zones) or 10 bars minimum (for ‘null’ data, if allowed?)

Total module width: 34 + 11n

Total module width, including quiet zones: 54 + 11n

Total number of bars: 10 + 3n

Data length <= (NumBars - 10 )/3

Possible Code 128 if (NumBars - 10) mod 3 == 0

Note: invisible characters (control characters and FNC1–4 count but will not show up in input or output text length)

shifts between three code sets (A, B, and C) to enable representing all 7-bit ASCII characters

Code sets A and B cover 128 characters; code set C represents digits more efficiently

Format:

1. Start symbol (A: 211412, B: 211214, or C: 211232; 211133 is ‘reverse stop’ symbol indicating barcode is upside down, so either way code always starts with ‘211’!)

2. Encoded characters

3. Check symbol (weighted sum, modulo 103, of all symbols up to check position)

4. Stop symbol (233111 or 211133 reverse)

5. Final bar (2 modules wide)

Character sets:

128A (Code set A): ASCII 00-95; 0–9, A–Z, control codes (except 127, ␡), special characters (<space> !”#$%&’()*+,-./:;<=>?@[\]^_), and FNC1–4

128B (Code set B): ASCII 32-127; 0-9, A-Z, a-z, ␡, special characters (from code A, plus `{|}~), and FNC1–4

128C (Code set C): digits 00-99 and FNC1 (each group of two digits is stored with one of 100 different symbol patterns)

If an odd number of digits is to be encoded, shift to Code A or Code B prior to the final digit

Code A and code B use the same symbols to encode characters common to both codes

FNC4 causes next character to shift by adding (or subtracting) 128 from the ASCII value. FNC4FNC4 acts as a ‘shift lock’ until that double code occurs again.

10 module quiet zone required at beginning and end of code

Shifting between code sets:

From A: code 98(shift->B), 99(->C), 100(->B)

From B: code 98(shift->A), 99(->C), 101(->A)

From C: code 100(->B), 101(->A)

FNC characters

Code A: code 96 (FNC3), 97 (FNC2), 101 (FNC4), 102 (FNC1)

Code B: code 96 (FNC3), 97 (FNC2), 100 (FNC4), 102 (FNC1)

Code C: code 102 (FNC1)

GS-128: Subset of Code 128; Formerly known as UCC/EAN-128

Two dimensional codes (linear, stacked)

PDF417

PDF417: Portable Data File with 4 bars (both light and dark, per symbol/codeword) and 17 units (modules) long); Symbol Technologies, c. 1991; ISO 15438

Rows: 3–90

Row hight at least 3 times minimum width (module width?)

One unit spacing between elements:
Base 929 encoding

900 codewords for data; 29 for special functions

Each pattern:

4 dark bars
4 light bars
17 units total
always starts with a dark bar
always ends with a light bar (i.e. whitespace)

Three sets of bar patterns for each codeword: 0, 3, and 6

these clusters are alternated between rows
this allows any single row to be identified to a ROW mod 3

Encoding modes:

Binary / byte: 6 bytes per 5 code words, then one byte per code word for excess bytes
Numeric: n digits encoded in [n/3]+1 codewords

Maximum 44 digits in 15 codewords

Text: Each codeword encodes 2 Base 30 digits, which represent printable ASCII (plus CR, LF, HT) with four submodes:

Uppercase: A-Z, SPACE, ->LOWER, ->MIXED, ->NextDigitPunctuation
Lowercase: a-z, SPACE, ->NextDigitUPPER, ->MIXED, ->NextDigitPunctuation
Mixed: 0-9, CR, TAB, SPACE, « &,:#-.$/+%*=^ », ->PUNCT, ->LOWER, ->UPPER, ->NextDigitPunctuation
Punctuation: « ;<>@[\]_`~!,:-.$/“|*()?{}’ », CR, TAB, LF, ->UPPER

Between 2–512 error detection and correction codewords added, using a Reed-Solomon ECC
51 square modules needed to represent 10 bits, compared to 8 modules for 8 bits with codes such as DataMatrix or QR codes, or 4 times as much area for the same amount of data (excluding metadata and space used for ECC)

Quiet zone, left

Start pattern: 811111121 (11111111010101000)

Left Row Indicator: 17 units; Encodes row number and ECC level

Data Codewords: 16 + 1 units each; Quantity: 1–30 per row

Right Row Indicator: 16 + 1 units

Stop Pattern: 18 units; 711311121 (111111101000101001)

Quiet zone, right

Two dimensional codes (square)

Aztec

Maxicode

QR code

Max characters, numeric: 7,089 (3.33 bits/char)
Max characters, alpha: 4,296 (5.5 bits/char)
Max characters, binary: 2,953 bytes (8 bits/char)
Dots per side (min 21): 17 + (4 x version number)

Character sets:

Numeric (10 characters): 0–9
Alphanumeric (44 characters): 0–9, A-Z, (space), $%*+-./:
Binary: ISO/IEC 8859-1 (LATIN-1)

4 error correction levels:

Level L: 7% data loss allowed
Level M: 15% data loss allowed
Level Q: 25% data loss allowed
Level H: 30% data loss allowed

9 Modes:

0000 | 00 : End of message
0001 | 01 : Numeric
0010 | 02 : Alphanumeric
0011 | 03 : Structured append
0100 | 04 : Byte encoding
0101 | 05 : FNC1 in first position
0110 | 06 : (unassigned??)
0111 | 07 : Extended Channel Interpretation (ECI)
1000 | 08 : Kanji encoding
1001 | 09 : FNC1 in second position
1010 | 10 : (unassigned?)
.... | .. : (unassigned?)
1111 | 15 : (unassigned?)

Two dimensional codes (rectangular)

Two dimensional codes (other)

References

TEC-IT