Thanks for visiting Imaging and Machine Vision Europe.

You're trying to access an editorial feature that is only available to logged in, registered users of Imaging and Machine Vision Europe. Registering is completely free, so why not sign up with us?

By registering, as well as being able to browse all content on the site without further interruption, you'll also have the option to receive our magazine (multiple times a year) and our email newsletters.

OCR aids mail sorting for TNT Post

Share this on social media:

TNT Post, the UK mailing arm of global parent company TNT, has implemented a second Optical Character Recognition (OCR) system to help handle the high throughput of mail at its UK sorting centres. The Address Reader from Netherlands-based Prime Vision, specialists in character recognition in postal and banking automation, will run in parallel to Prime Vision's MMT Sabre address recognition system already in place.

In the past 12 months, the company has experienced more than a 60 per cent growth in mailing volumes and now delivers an average of 160 million items a month. The two OCR systems provide automated mail sorting, with the Address Reader capturing handwritten information and processing address images that are not picked up by the MMT Sabre OCR engine.

Chris O'Malley, head of sort automation at TNT Post, commented: 'We expected our Sabre OCR to read from 90-95 per cent of our mainstream, machine-printed addresses and this has now been proven. We also knew that the incorporation of Prime Vision would have the potential to increase this read rate. In fact tests showed an improvement of up to 4.24 per cent when using both OCRs. This, together with the potential to develop our ability to sort handwritten mail automatically prompted us to install a PV Address Reader on each of our UK sorting machines.'

O'Malley continued: 'We've done some handwritten mail trials with the untuned and basic PV Address Reader. Even at this stage we are getting more than 56 per cent read rate. We are now collating written images from all our sorting centres so that the development work can start in earnest.' Also being evaluated is the possible addition of video coding.

Recent News

26 September 2019

Rugby fans are now able to watch highlights from the Rugby World Cup, currently taking place in Japan, from angles and viewpoints not possible with conventional cameras, thanks to a multi-camera system from Canon

13 September 2019

A hyperspectral imaging system built by US research centre Battelle, using Headwall sensors, has been chosen as a finalist for the Department of Homeland Security’s Opioid Detection Challenge

23 July 2019

On the 50th anniversary of the Moon landing on 20 July 1969, Zeiss has described how, in less than nine months, it built the camera lens used to capture the iconic images during the Apollo 11 mission

18 July 2019

Researchers at Lund University in Sweden are using high-speed cameras to study how insects use visual information to control flight