Google’s new patent to index text in video and photos

Tech Crunch recently posted an article highlighting that in July of last year, Google filed a patent application on image and video search technology. However it was only recently that it made the news public that it plans to patent technology that will allow its search bots to understand specific text featured in images and video clips.

This has the potential of massively changing the search landscape, especially for websites that have a lot of photos or links to video content. Google's new image search technology would mean that photographs and videos indexed by Google would be searchable by the text located within them and presents a huge step forward in the indexing sites that have not previously been available for search.

Digital images often do include text, but were previously unavailable for search since it was impossible to index using older search technology. Image text (text inside an image) typically includes text of varying size, orientation, color and typeface.

For example, text in a digital image derived from an urban scene often provides information about the displayed landscape or specific location. A typical street scene would include text as part of street signs, building names, physical address numbers, window signs and a lot more.

This technology could probably make every book in the Google Books database really searchable, with the next step being YouTube, Flickr, Picasa Web and more.

It will be interesting to see how Google develops this new image search technology.