2D code - the next step for bar codes

With the small data capacity of bar codes increasingly limiting their usefulness, Omron's Sean Eley looks at the emergence of the two dimensional code, and the equipment to print and detect it

With the small data capacity of traditional bar codes increasingly limiting their usefulness, Omron's Sean Eley looks at the emergence of the two dimensional code, and the equipment to print and detect it.

The humble bar code has been meeting data portability requirements industry-wide for decades, unmatched in its ease of application and of data extraction, and untouchable in its cost-effectiveness.

It is not without its detractors, who point out the fragility of the print quality in aggressive environments and the implications for quality control, and who champion the applications for read/write technology, but still the bar code marches on largely unchallenged.

It is true that the emerging read/write technologies have carved out niches for themselves in a number of markets.

RFID tagging for one is a mainstay in automotive production, and is coming to be used increasingly in food production.

We are reaching the stage where the pricepoint of these tags is such that, in some areas, they can be regarded as disposable technologies.

In the airline baggage handling market, for example, the latest RFID tags are coming to be seen as an ideal means of tracking luggage for a low implied cost.

Indeed, Omron itself has launched tags into this 'disposable market' at a pricepoint of below 50 cents for volume production.

But this is still a long way away from the cost/performance panacea of the bar code.

It may be a read-only technology, but in numerous market sectors, that is exactly what is required.

However, it is becoming apparent that the bar code we know and love is starting to show its age.

In this information society, the amounts of data we try to carry around on products have been escalating wildly, reaching and exceeding the limits of the bar code's capacity.

At the same time, as data requirements rise, so does the need for providing error checking on that data - something again to which the traditional bar code has no answer.

What is needed is a read-only technology with all the virtues of the bar code but with vastly increased capacity.

And so a new challenger is emerging - the 2D code.

Whilst a conventional bar code encodes information in one direction only, the 2D code extends this with a matrix across both the X and Y axes.

For the standard bar code, the only way to add more information is to greatly extend the length of the bar code, making it much more difficult to read.

At the same time, there is no mechanism for redundency, and a very high contrast (80% or more) is required to ensure accurate reading of the code.

As a result, it doesn't take much degrading of the print of the bar code before it becomes unreadable.

By contrast, the 2D code, offering expansion in two directions, can hold a much higher volume of information.

In addition it provides redundency by error checking, and, impressively, requires only a 20% contrast to ensure accurate reading, so misreading is almost impossible.

High density For a given 'area' of code, where the bar code can hold say 20 bytes of information, a 2D code of the same area can contain some 2000 bytes, in a combintation of alphanumeric and binary forms, with data compensation.

As a high density format, the 2D code allows either the size of the mark to be reduced for a given amount of information, or for a greater level of information to be held without increasing the size of the mark.

With its capacity for increased levels of information to be encoded, the 2D code lends itself to 'track and trace' applications.

And with its capability for error checking and correction, and its ability to be easily printed and read even at low contrast levels, it is an inherently safe form of coding.

In addition, against the subjective blandness of the bar code, there is a certain aesthetic appeal about the 2D code.

There are two types of 2D codes.

Stack types are more obviously derived from the standard bar code, and can be readily recognised as a bar code expanded in two dimensions.

But alongside the stack types are the matrix types, and it is these which are proving to be the most useful.

Two in particular are moving towards becoming defacto standards, with the 'data matrix' code becoming a standard across Europe and the US, and with the 'QR' code gaining acceptance in Japan.

Of the two, the data matrix code has the advantages of smaller size and higher safety in reading.

A key advantage of the data matrix code, however, from the user's point of view, is that it is extremely easy to print.

The user can select from either round dots or square cells, and can employ either a square or rectangular matrix.

Any type of output device can be used to print the code, from inkjet, laser etch and dot matrix printers, to thermal transfer, web press and laser printers.

The qualities of the data matrix code mean that the stringent print quality requirements of traditional bar codes can be relaxed.

As discussed, only a 20% colour contrast is required, and cells can be printed light-on-dark or dark-on-light.

Cell colour can be based on a single pixel or multiple pixel averaging, and the format offers compatibility for normal image or mirror image printing.

The format also assures ease of reading.

Continuous orientation bars around two sides of the mark provide information for orientation.

Alternate light and dark cells around the remaining two sides define the number of rows and columns.

A further mark gives identification data for the centre of the cell.

Having registered all this, the pattern decoder can then set about extracting the coded data.

As the data matrix code has evolved, the error correction capabilities have greatly improved.

It is now the only symbology to offer both Reed-Solomon and Convolutional code error correction.

Reed-Solomon is a byte correcting scheme preferred for block damage recovery, whilst Convolutional code is a bit correcting scheme preferred for random damage recovery.

Importantly, the technology to read and decode the 2D mark is both affordable and easy to use.

Omron has recently launched a system built around its F150 visual inspection system, but with a dedicated operating system and tailored software for 2D code reading applications.

The resulting V530-R110 system is an extremely compact solution for reading both QR and data matrix codes, combining high performance with ease of use, and at a fraction of the cost of a traditional vision system.

The system is easily programmed via on-screen menus, with all selections made using a hand-held keypad.

At a cost which makes using 2D codes viable for just about any application, the V530-R110 system meets the requirements of 2D code reading by combining pattern matching with greyscale inspection.

The system has the capacity to store numerous image templates, so it can readily differentiate between different 'flavours' of QR and data matrix codes, with their implications for size of the mark, error correction technique, print style and information content.

In use, the system is first set up to detect a particular variant of 2D code.

When the mark passes beneath the system's camera, the system first registers that a mark is indeed present (in whatever orientation) and then looks for the orientation marks using its pattern matching capabilities.

With these found, the system can go on to identify the numbers of columns and rows and the error correction methodology.

With all this ascertained, the system then switches to extracting the coded information from the mark.

Data can then be passed to a terminal over an RS232 link as readable and meaningful information.

With its advancement of the cause of the bar code, 2D coding looks set to become the portable information medium of choice in a whole host of applications where the limits of the conventional bar code have been reached.

In particular, in sectors where technologies such as RFID tagging still have too great an implied cost, the 2D code at last offers a viable, high performance alternative.

Back to top