Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Understanding CAN/CSA-ISO/IEC 17877-04: PDF417 Bar Code Symbology Specification
Technical Overview of the International Standard for Two-Dimensional Bar Codes in Automatic Identification Systems
Scope and Introduction
CAN/CSA-ISO/IEC 17877-04 is the Canadian adoption of the international standard ISO/IEC 17877:2004, titled Information technology — Automatic identification and data capture techniques — PDF417 bar code symbology specification. This standard defines the physical, optical, and data encoding requirements for the PDF417 two-dimensional bar code symbology. PDF417 is a stacked, continuous, multi-row symbology used extensively in logistics, transportation, identity documents, and healthcare for encoding large amounts of data reliably.
Key application areas: PDF417 is used on driver’s licenses, parcel labels, airline boarding passes, and medication packaging, offering high data capacity and error resilience.
The scope of this standard includes the complete specification of the symbology: symbol structure, codeword assignment, data encoding rules, error correction algorithm (Reed–Solomon), physical dimensions, print quality parameters, and decoding requirements. Compliance ensures interoperability between encoding systems and scanners worldwide.
Technical Requirements
Symbology Structure
A PDF417 symbol consists of a stack of rows, each row containing a start pattern, a left row indicator, data codewords, a right row indicator, and a stop pattern. Each codeword is represented by a set of bars and spaces of varying widths (module-based). The standard defines three compaction modes for efficient data encoding: Text, Byte, and Numeric.
Data Encoding
The encoding process begins by selecting the appropriate compaction mode(s) based on the data type. Data is converted into codewords (values 0–928) using a base-900 conversion. The standard defines a sophisticated switching mechanism between compaction sub-modes to maximize density.
Error Correction
PDF417 employs Reed–Solomon error correction codewords appended to the data. The standard defines nine error correction levels (0 through 8), where level 0 offers no correction and level 8 provides the highest resilience. The following table summarizes the relationship between error correction level and the number of error correction codewords:
| Error Correction Level | Number of Error Correction Codewords | Typical Application |
|---|---|---|
| 0 | 2 | Minimum, controlled environments |
| 1 | 4 | Low security |
| 2 | 8 | General use |
| 3 | 16 | Standard logistics |
| 4 | 32 | Transportation |
| 5 | 64 | Healthcare |
| 6 | 128 | Identity documents |
| 7 | 256 | High security |
| 8 | 512 | Maximum resilience |
Note: The total number of data codewords is reduced as error correction increases. The standard provides tables for selecting the combination of rows and columns to achieve the desired capacity and robustness.
Dimensions and Tolerances
The standard defines the nominal module width (X-dimension) and row height. The minimum X-dimension is 0.265 mm (0.0104 in) for printed symbols, with recommended tolerances for quiet zone (minimum 2X on each side) and pitch. For scanning compatibility, the row height should be at least 3X. Symbol size can range from a few rows/columns to maximums of 928 codewords.
Important: Quiet zones must be free of any printing marks or borders. Violation of quiet zone requirements is a common cause of read failures in line scanners.
Implementation Highlights
Successful implementation of PDF417 requires careful attention to both encoding and printing processes. The standard includes specifications for print quality (grade based on ISO/IEC 15416) and symbol contrast. Decoders must be capable of reading symbols at any orientation due to the omnidirectional scanning ability of PDF417.
Tip for integrators: Always select an error correction level appropriate for the anticipated symbol damage. Level 4–6 is recommended for most logistics applications; level 6–8 is preferred for ID cards and high-security documents.
Data capacity varies with the number of codewords and error correction level. In Numeric compaction, up to 2710 digits can be stored in a single symbol. In Text compaction (alphanumeric), up to 1825 characters, and in Byte compaction, up to 1100 bytes. Hybrid encoding using multiple compaction modes can further optimize density.
Advantage: PDF417 does not require a database lookup — it is a portable data file. The symbol can contain complete information, making it ideal for offline verification.
Compliance and Verification Notes
Compliance with CAN/CSA-ISO/IEC 17877-04 is typically assessed using calibrated symbology verifiers that measure parameters such as minimum reflectance, symbol contrast, modulation, defects, and decodability. The standard references ISO/IEC 15415 (two-dimensional print quality) for grading.
Non-compliance risks: Symbols printed without adherence to the specified X-dimension, quiet zone, or error correction algorithms may be rejected by verification systems and cause scanning failures in automated identification networks.
Certification programs for PDF417 printers and readers are operated by standards bodies and industry organizations. Conformance testing ensures that data encoded per the standard can be decoded correctly and that error correction recovers the intended information even when portions of the symbol are damaged.
This standard is part of a family of bar code symbology specifications under ISO/IEC JTC 1/SC 31. For detailed implementation, refer to the normative sections of the full document and supporting standards on bar code print quality and scanner performance.
Frequently Asked Questions
Q: What does PDF417 stand for?
A: PDF stands for Portable Data File, and 417 indicates that each codeword is constructed from 4 bars and 4 spaces (total 17 modules). PDF417 is a stacked two-dimensional symbology that can encode large amounts of data in a small space.
A: PDF stands for Portable Data File, and 417 indicates that each codeword is constructed from 4 bars and 4 spaces (total 17 modules). PDF417 is a stacked two-dimensional symbology that can encode large amounts of data in a small space.
Q: How does Reed–Solomon error correction work in PDF417?
A: The standard uses a systematic Reed–Solomon code with 8-bit symbols over GF(929). The encoder appends a set of error correction codewords calculated from the data. The decoder can correct both erasures and errors up to the capability of the selected level. Level 8 can correct up to half the total number of error correction codewords in error.
A: The standard uses a systematic Reed–Solomon code with 8-bit symbols over GF(929). The encoder appends a set of error correction codewords calculated from the data. The decoder can correct both erasures and errors up to the capability of the selected level. Level 8 can correct up to half the total number of error correction codewords in error.
Q: What are the typical data capacity limits of a PDF417 symbol?
A: The maximum data capacity depends on the number of rows and columns and the error correction level. In Numeric compaction, up to 2710 digits; in Text compaction, up to 1825 characters; in Byte compaction, up to 1100 bytes. Higher error correction reduces data capacity.
A: The maximum data capacity depends on the number of rows and columns and the error correction level. In Numeric compaction, up to 2710 digits; in Text compaction, up to 1825 characters; in Byte compaction, up to 1100 bytes. Higher error correction reduces data capacity.
Q: What is the difference between PDF417 and other 2D barcodes like Data Matrix or QR Code?
A: PDF417 is a stacked linear symbology (multiple rows of linear bar patterns), whereas Data Matrix and QR Code are matrix (square) symbologies. PDF417 generally offers higher data capacity and is optimized for vertical scanning, while matrix codes offer better performance in omnidirectional scanning and smaller symbol sizes. The choice depends on the application and scanning environment.
A: PDF417 is a stacked linear symbology (multiple rows of linear bar patterns), whereas Data Matrix and QR Code are matrix (square) symbologies. PDF417 generally offers higher data capacity and is optimized for vertical scanning, while matrix codes offer better performance in omnidirectional scanning and smaller symbol sizes. The choice depends on the application and scanning environment.
This article provides an overview of CAN/CSA-ISO/IEC 17877-04:2004. For official verification and certification, always consult the complete standard document. Last updated 2026.