Welcome to WebmasterWorld Guest from 22.214.171.124
Forum Moderators: goodroi
A patent application filed by Google (NSDQ: GOOG) with the World Intellectual Property Organization in June 2007 and published on Thursday points to more sophisticated search techniques.
The filing suggests that the privacy issues raised by Google Maps Street View will only get more complicated, that YouTube searchers may one day be able to conduct keyword searches for text captured on video, and that Google searches may one day return a list of products on local store shelves.
"Recognizing Text In Images" is an application to patent a method of optical character recognition in digital images.
Google Patent: Recognizing Text In Images [informationweek.com]
Methods, systems, and apparatus including computer program products for using extracted image text are provided. In one implementation, a computer-implemented method is provided. The method includes receiving an input of one or more image search terms and identifying keywords from the received one or more image search terms. The method also includes searching a collection of keywords including keywords extracted from image text, retrieving an image associated with extracted image text corresponding to one or more of the image search terms, and presenting the image.
RECOGNIZING TEXT IN IMAGES [wipo.int]
That recognising something certain that you are looking for though, not text in an image which could be anywhere, broken in anyway, different colours, font types yada yada...
joined:Apr 25, 2002
It's great marketing though. By filing patents frequently for technologies that probably aren't even close to workable in the near future, Google appears to be a technology leader and innovator without having to deliver anything and without even promising anything at all. It's much more savvy than the vaporware strategy that got MS so much bad press.
joined:Oct 27, 2001
I can envision a whole new generation of link spam:
Graffiti artists are now the equivalent of forum spammers.
Those flyers stapled to telephone poles at traffic lights are now the equiv of junk mail.
The homeless guy with sign is now a pop-up ad. (a dhtml one at that!)
Writing "I will not ..." 100x on a chalkboard is keyword stuffing.
I think Iím going to rock salt my back yard with keywords and hyperlinks.
Anyone want to buy a link?
(Just kidding G)
I don't know that these new technologies aren't close to workable...aren't all the video search engines (Blinkx, Pixsy, Everyzing, Truveo) in the business of searching videos using both image recognition software and voice recognition software? That is why those schooled in video optimization recommend using titles on the video screen, etc. So, the question for me is how much is the technology that Google is trying to patent like that image recognition software used by video search engines?
Google's Lab experiments aren't always what they initially seem to be. Google Catalogs wasn't about mail-order catalogs, it was about Google having a test set to hone their OCR technology so they can apply it to everything.
joined:June 15, 2001
A camera takes a picture. There are hundreds of cameras on the market. Most if not all have 1 or more patents on them. It could be something as stupid as something to deal with the lens cap or something as unique as a new type of image format. It doesn't matter, as long as it is different from past inventions in some manner.
So a person wanting to buy a new iPod would take a picture of the ipod, the technology would recgnoise the item and search a database of registered retailers and order 1 from the cheapest matching product.
The iPod would then be delivered to your doorstep. Obviously, to use the service you would be registered in a central database that either had funds already deposited or was linked to your credit card/bank etc.
I can't remember where I seen this, I might have read it in the newspaper or seen a report about it on tv. Although I do not know if it would be exactly the same, I can see it being very easily adapted if a mobile phone contained a bar code scanner and that would be able to search much more easily and much less possibilities for ordering incorrect items.
There have been a ton of patents granted with regard to link analysis (or text recognition or semantic analysis) in search, but while they're all analyzing for the same purpose, the process and data being used, and the particular way they're being applied, are different.
There was a thread here at WebmasterWorld way back in 2001 about Alta Vista having just about everything for search patented, but that didn't stop any further patents from being issued to other companies.