ISO/IEC 15416 Barcode Quality Specification
Wednesday, January 06, 2010
Posted in: Verification
ISO/IEC 15416 Barcode Quality Specification: Hype, Reality and Good Faith
The ISO spec for quality of linear barcodes is one of the most often quoted, misquoted and misunderstood documents. Reminds me of the Bible: quoted, misquoted and misunderstood and probably for the same reason: most people don’t really read it. That’s a blog for another venue and day. For example, two so-called bedrock parameters for ISO 15416 are the ten-scan average and the 660nm wavelength requirement for light source. Neither of them is hard-wired into the ISO/IEC 15416 specification.
The main text of the ANSI/CEN/ISO specification does indeed state that the Overall Symbol Grade is based on a ten scan average, but the appended language states that when there is low possibility of defects in the barcode the number of scans per symbol can be reduced to as few as 2. Refer to Annex G and J for the exact language.Then there’s the requirement that verification be done in 660nm red light. Actually the ISO specification does not require any particular wavelength of light for verification. What it says is that verification should be performed in the same wavelength as the scanner that will ultimately decode the symbol. Certainly most of the scanners out there are built with 660nm + 10nm but not all of them are. The wavelength is specified in the industry application specification—not in the ISO/IEC 15416 specification. Thus it is not always wrong to verify in some other wavelength, such as white light.
These apparent discrepancies do not discredit ISO/IEC 15416. The specification knowingly attempts to do the impossible: to nail down variables in a rubber world. Some of these variables just won’t be nailed down—and in addition to that, scanning technology is evolving and improving. The specification is meant to give shape and direction to barcode quality and to make barcode symbol performance predictable and repeatable.
Here’s one example of a verification system that deviates from ISO/IEC 15416, 
the rationale for doing so, and how to control it.
We have installed automated, unattended verification on high speed variable printing systems.
By “variable” I mean sequenced barcodes, not repetitions of the same encoded data.
Lighting was a major concern for two primary reasons:
1. Ambient light was intense and variable, and not optional. The work area had to be well illuminated. In the work area of one production floor the overhead lighting was high-intensity sodium; in another it was mercury vapor. Both lighting systems produce a lot of red spectrum. Verification lighting needed to overcome that. In a third situation, the shop floor was bathed in fluorescent light. This in combination with the high speed produced false light and dark readings because of the pulsing of the fluorescent lights. Again, verifier lighting had to overcome this. White light was the best available technology.
2. The high speed nature of the situation also required very intense light.A third factor helped mitigate the decision to use white light: the barcodes were always black on white, so color contrast was never going to be an issue. Nevertheless we felt it was important to correlate the reflectance values and contrast grades achieved by our white light system to a calibrated and ISO conforming 660nm verifier. In all cases the customer’s test station verifier was used as the reference standard. First we re-calibrated using a Certified Conformance Standard Test Card to make sure their verifier was in specification to ISO/IEC 15416, and then we made our high speed automated system match it as closely as possible. In all cases we got well within the tolerance for conformance of their lab verifier.
This is not bending the rules—this is making use of flexibility built into the rules—even ones that are considered “bedrock”. It is not always possible to follow ISO/IEC 15416 (or ISO/IEC 15415 for verification of 2D symbols), whether for technical or economic reasons. Good faith must always be the cornerstone. The ultimate goal is to produce high quality, high performing barcodes that meet the requirements of the customer and the marketplace.In some applications and industries the requirements and methodology are well thought through, clearly spelled out and usually achievable. But sometimes even such clearly defined requirements are poorly thought out and unrealistic. And at other times the verification requirements are ambiguous or even nonexistent.
Again, the goal is to meet the needs of the customer and the market. If one is honest about the methodology, the symbols meet the expected quality level and the customer is served. Then and only then can the verification process be considered appropriate and successful.
