OCR (Optical Character Recognition) is one of most explored applications of Computer Vision. It's possible to find lots of OCR software and apps for mobile devices.
In a fast search in Google Play, there are dozens of options for test in our Android devices. Some apps were developed to be used as scanners of text, translators or helping tool for learning to read.
Blind people needing to read a label in a box, turists in a foreign country or children learning to read, those and many other public would be much interested in OCR solutions embedded in a mobile device.
In fact, OCR has a great potential to be used in many kinds of solutions, specially for mobile devices.
OCR uses images to make the recognition and these images can be from some sources, for example camera or video streams.
Presented examples, scanners of text, translators and helping tool for learning to read, use images from camera, but it's possible to apply ocr algorithms in video streams and "read" text from a video that is stored in a server and consumed on demand by the device.
OCR can be used as basis for more complex solutions, like augmented reality.
In industry, OCR can be used of many ways, but automation is probably the most impacted area with OCR solutions.
Using OCR features in robots, machines can identify products, read values in displays of sensors, follow written instructions and do many other tasks.
In this post, my intention is to share some knowledge and a way of thinking about OCR, openning the perception of how this technology can help lots of people.
Take a look in existing solutions and search for non-solved problems, think if a mobile device was able to analyse realtime images this problem still happening.