Optical Character Recognition (OCR) and Integration with PhotoRobot

With PhotoRobot, creating high-quality, detail-rich product content is easy, but what if you need to capture text on a photographed object?

With PhotoRobot, creating high-quality, detail-rich product content is easy, but what if you need to capture text on a photographed object? This is where Optical Character Recognition (OCR) comes into play. No matter if you’re capturing product spins, stills or planograms, the PhotoRobot_controls software suite has you covered with OCR support for all of your imagery. This tool comes at no extra cost, and it can extract text directly off of items to include in your export feed and easily upload to webshops or product management tools. All you need is time for a quick review before publication to the web!

PhotoRobot's Optical Character Recognition

With online product content, there is high demand for not only quality product photos but also text related to the product itself -- this can be used for captions, quick search features, and streamlined indexing into a product portfolio, which is made easy thanks to PhotoRobot Optical Character Recognition (OCR) Integration. Some online marketplaces even require labels to be presented in writing, which provides us with another considerable reason for digging deeper into automation for text recognition.

The PhotoRobot solution is powered by many Google cloud technologies, and it’s here Google Cloud’s Vision comes to the scene with its extremely powerful pre-trained machine learning models that assign labels to images and quickly classify them into millions of predefined categories. 

With this solution, you can detect objects and faces, read printed and handwritten text, and build valuable metadata into your image catalogue. Our OCR module is based on functions related to the printed and handwritten text functions.


How does PhotoRobot’s OCR module work?

As the text field is frequently split into more images of the product, PhotoRobot_controls combine more images together automatically and then flatten the text in a way that makes it easy to recognize the texts in logical groups and then provide you with an extracted ready-made text source four your final inspection before online publication. 

The human operator has a product spin available online, so this becomes very much like holding the product in hand. Texts are immediately extracted and ready for pre-publication review. The optimized output is then a part of the data which is exported alongside the images using our standard dynamic export feeds. 

A great advantage to this is in the fact that the operator can work with everything remotely, meaning even freelancers and home-office workers can assist with this process. 

A license covering general output supervision and performing OCR costs less than €2 per day, so what are you waiting for? Contact PhotoRobot experts today to help you find the best setup for your zero-friction imagery production.

Published:

22.9.2020