Case Study

Barcode Reading

We were approached by an existing client, a dominant international postal and logistics organisation that were experiencing increasing frustration with failed Barcode Reading. The successful read-rate was almost 95%, but the 5% fail-rate was having a significant effect on parcel volumes.

For every failed barcode reading, manual intervention was required to avoid any potential bottlenecks. Consecutive unreadable labels had a knock-on effect that resulted in excessive delays contributing to sometimes missed deadlines and possibly broken contractual agreements. Of particular concern, the operators whose job it was to manually process these parcels where always under a constant pressure to decipher the failed label details as quickly to avoid these costly problems building up.


One considerable problem lay in that industry-wide systems were designed with no automated response for failure. Simply put, these sortation engines were never configured for attempted re-reads, leaving these manual interventions the only option to keep the parcels flowing.

There could be many reasons for a failed read, but in fact, it came down to two main issues. The erratic label standards and ironically, the parcels themselves. By nature, packages come in all shapes and sizes, often with curved edges and surfaces restricting the label being presented flat to the scanner. Often parcels are wrapped in plastic which can render the label partly invisible.

Concerning the labels, the most significant problem is that often barcodes don’t meet industry specifications. Printed barcode labels are inconsistent in quality; sometimes lines are blurred, uneven, incomplete or damaged.

The quiet zones can be too small or the bar ratios infrequent. Also, low contrast, image blur and poor printing can lead to binarisation issues, literally leaving the barcode readers unable to tell black from white.

Any of these problems can produce the same result – a failed read. And every-time a human is forced to get involved it becomes costly. A 5% failure-rate means one-in-twenty parcels manually processed. With millions of deliveries sent every day, that’s a severe amount of wasted time and money.


The R&D team at Prime Vision set about defining the challenges that faced them. They needed to develop a system that could read any label no matter the circumstance. The project was led by Senior Researcher Sjaak Koomen. “We’re considerable proud of this innovation. The key to the solution lay in our approach. Rather then re-designing the bar-coding equipment; we changed the way the different components interact, and crucially, how they thought. We built our solution on the assumption that all printed barcodes did not comply with industry specifications. We based our software on the barcode material experienced in real life, and then by utilising a neural-network and applied machine intelligence, we have created a solution that can learn to correlate a deformed signal to an ideal signal.”

Of particular note, one impressive innovation was in the handling of blurred images. In a painstakingly complicated process, the R&D team set about creating artificial black-and-white barcode signals and then manipulating them to simulate every possible distorted signal that could result from a distressed label. Once paired, these signals are used to create a neural-network, enabling machine intelligence that can learn to correlate a deformed signal to an ideal signal.

In simple terms; when the system is presented a disturbed signal from the parcel, it says to itself. “If I can see such a disturbance, then I know this must be the sharp crisp ideal Signal”. The result effectively eliminates the blur for any barcode.


Sjaak and the team are genuinely proud of this innovation and talk confidently of the benefits, “Overall, the new solution is capable of reading 70% of the previously failed reads, taking the successful readrate from 95% to 98.5%. Putting these figures in perspective; a smaller National Parcel Carrier will
typically process more than one million parcels a day. That’s 35,000 parcels that would have to be manually processed if our software hadn’t been able to read them”.

This makes business benefits clear. Even by allocating just 30 seconds to manual process a failed read, this solution saves 291 working hours to process the 35,000 failed reads. Per day. The more barcodes read, the higher number of parcels remain in the pipeline to be sorted automatically resulting in less manual intervention and faster processing; which is ultimately saving money.

Just as significant, are the benefits to the equipment operators. The higher rate of successful reads considerably reduced the pressure on this potential bottleneck. The reduced failure rate leaves over 300% more time to process the remaining failed reads at the same operating costs, reducing pressureinduced secondary errors from hurried operators.

With the project now fully operational and the client enjoying the benefits, Sjaak considers the project as a huge success. “It’s remarkable that all these positives come from one solution. A solution consisting of small innovations. But it’s these small steps that made the real difference. Together they add up to the 3.5% read gain. 3.5% gain sounds small. It is small, but it’s a substantial gain at the really tough end of the equation. It’s the 3.5% that counts, and of course, it’s the 3.5% that can turn the loss to profit”.

Get a free consult

    Privacy Preferences
    When you visit our website, it may store information through your browser from specific services, usually in form of cookies. Here you can change your privacy preferences. Please note that blocking some types of cookies may impact your experience on our website and the services we offer.