What about indexing the jpeg tags such as location, camera and title/subject if they are included? There is also often copyright info embedded in the file. You can see this if you open one in a hex editor. All of this should be readable and indexable.
I made a search query and got a result with the blurb matching only the text in an image. Just to be sure i went to google's cache of it and it still only matched the image.
Too bad most of the time the image is an advertisement.
There's also text within the images themselves such as their headers. Similar to the way camera images contain EXIF information? Will that get spidered too?
Matt will occasionally open up a Google Moderator page for questions and announce it on his blog and on Twitter. Generally, your best bet for asking specific questions is in the Google Webmaster Help Forum. We have a link to it on our main channel page. (Click on "GoogleWebmasterHelp" at the top of this comment to get there.)
Have faith my fellow developers; everything is possible.
AbartsWorld 1 year ago
What about indexing the jpeg tags such as location, camera and title/subject if they are included? There is also often copyright info embedded in the file. You can see this if you open one in a hex editor. All of this should be readable and indexable.
hostroute 1 year ago
its not too hard , Google can use neural networks application and AI which can recognize character in images .
1111l 2 years ago
They are already doing it.
I made a search query and got a result with the blurb matching only the text in an image. Just to be sure i went to google's cache of it and it still only matched the image.
Too bad most of the time the image is an advertisement.
kodos84 2 years ago
perhaps this image is showing because it has links going to it from other sites, and these are text links.
thestretch 2 years ago
Taking photos of nearly every street in America is a lot of work, too. But Google does it with street view.
DragonMasterMarf 2 years ago
Google will start to read text in your images by the year end.
Lol I'm just dreaming. But it is Google. Dreams come true... ;)
rahnas 2 years ago
There's also text within the images themselves such as their headers. Similar to the way camera images contain EXIF information? Will that get spidered too?
chasleo 2 years ago
At least one pdf I know of is returned on a query for a word that is only contained in a scanned image within the pdf. Looks like ocr to me.
iStylr 2 years ago
From what i knowed google ALREADY read text from images..
cedivad 2 years ago
No, they can only read the alt attribute from images.
metallica1987arg 2 years ago
It's cool how you know more about Google than the guys who work there.
oneeyedgeek 2 years ago
Lot of work for Google !
Maybe microsoft and yahoo have something to build upon then
almightyvegeta87 2 years ago
The pool table is the perfect surrounding for this kind of creative energies, isn't it? :)
sfussenegger 2 years ago 3
Is there other Google department using youtube to answer people's question? Like GMail?
TiriPon 2 years ago
We already have this in Scalable Vector Graphics, at least for computer generated graphics.
BTW, where can we submit questions to Matt?
nintendo9231889 2 years ago
Matt will occasionally open up a Google Moderator page for questions and announce it on his blog and on Twitter. Generally, your best bet for asking specific questions is in the Google Webmaster Help Forum. We have a link to it on our main channel page. (Click on "GoogleWebmasterHelp" at the top of this comment to get there.)
- Wysz, Google Webmaster Central
GoogleWebmasterHelp 2 years ago
"lot of work" = cost a lot of money and use up a lot of servers : )
woodsandhillsplc 2 years ago