Today Microsoft has put their optical character recognition (OCR) library for Windows on NuGet for developers to download. This library isn’t new but it is now much easier to access and include into projects. From Visual Studio, developers can search for Microsoft OCR at Project | Manage NuGet Packages. This library is free to download and there are no license fees associated with using the library in a project.
OCR can have a wide variety of uses in apps. Microsoft OCR scans images from a devices camera, local storage, or network storage for layout and text. Common elements such as email addresses, phone numbers, and URLs can be interpreted and acted upon in the app. This could be useful for scanning business cards, old contact cards, information from print ads, and more.
There are already examples of OCR apps on Windows and Windows Phone. Microsoft OneNote has had the ability to copy text out of photos for years now. Simple check-splitting app DivvyUp also uses OCR to read the bill and divide it accordingly. Being able to read and act upon hard-to-get information in print is becoming easier and easier.
Implementing Microsoft OCR is simple and easy to do. Microsoft gives some sample code on their Building Apps for Windows blog. One line of code calls the OCR engine and stores the results in a variable. From there the developer can choose what to do with the scanned text. Running it through an algorithm to find phone numbers, or emails is now fast and easy to do.
