Welcome to WebmasterWorld Guest from 220.127.116.11
Now what I think is the reason, is the hotlinking to my images, with those hot linking is also "some bad text" on those sites, also the art of sites could play a role.
Those hotlinking images is now in my site:domain.com search results on google image, means a lot of other domains show up, where i have no influence on there content, but they are damaging my Image theme so google think my site is bad, now thats my theory.
Its incredible how many ways another site can hurt your site.
Does anybody know how to do that?
Unfortunately, Google Image Search does seem to have a problem dealling with hotlinking, and that means your site suffers.
Here comes the flip side, I think your site can also benefit from Hotlinking.
Ranking Images for Web Image Retrieval
USPA 20080097981 awarded to Microsoft on 2008-04-24
That above patent provides some indication that hotlinking may be a benefit for some sites. I would think it could be used to do the exact same opposite as in the OPs example.
Once again a way another website can hurt your rankings, also the problem is not a small, I would say 99% dont think its necessary to change "moderate search off" to look at nature images or other normal images.
I also created a new site when google had, maybe still have this 302 redirecting bug, www or non www, hijacking problems, it was the best thing I did it really saved my business, so I think its time to create a new site again, once again be cause google dont focus on there search tool anymore.
bumpski: This may seem hard to believe, but I've found the sites you link to can prevent your images being indexed for the safe search filter.
I spent several months removing links to sites with any remotely unsafe keywords and found after 3 more months that my images were now being indexed by safe search.
Also make sure you don't link to a site whose images do not show up in safe search, even if there is no apparent reason.
tedster: If the site publishes user-provided images, there may accidental occurences of incorrect (and adult) meta data embedded in some or many of the files. Depending on how the user works with their software, incorrect metadata can happen. Enough accidents like this and the entire domain's images might be filtered.
Also, Google recently entered a patent application that offers a lot of clues on how images can be automatically processed for search. Here's the patent application [appft1.uspto.gov]. Notice these possibilities, especially when there is little or no data/metadata directly associated with an image:
- Images can be auto-tagged according to shapes, colors, and textures. This may involve breaking down images into smaller tiles and tagging those tiles.
- Images can be compared to other indexed images from around the web that have similar extracted features. Then keywords that are semantically related to those other images may be imported and used to tag the image that is being classified.
I can see lots of ways that an image might be incorrectly filtered. When it comes to image search, any search engine might prefer to err on the conservative side.
So while hotlinking may be part of the picture, there are other areas to consider when you run into Image Search troubles.
If such links do generate some kind of "adult by association", then that would be a rather grim flaw in the algorithm. It would allow other webmasters to trash your rankings - at least on image search.
However, if hotlinking to your domain is causing a problem in Google Image Search (and we are not 100% sure that is the case) then redirecting, substituting images, denying access -- none of that would change the fact that there's a link out there, pointing to your site that Google can spider.
Cloaking however would I think - have 2 sets of identical images, one for Google and one for everyone else; these are not cloaked and have separate urls. You then have 2 sets of pages, 1 for Google and 1 for everyone else; these are cloaked and you serve whichever to your visitors depending on whether you think they are Google or not.
If you serve Google's spiders pages which reference images that are never shown to human visitors, this would get around the problem as I understand it.
As far as Google is concerned the clean images are only ever referenced from your own domain. Hotlinkers would be referencing images which would exist if Google queried them directly, but would not be the ones that Google sees on your site.
A visitor would see the google image when viewing it on Google Images, but get the human version if they clicked through to the page.
If hotlinking is the problem of course, and assuming that the hotlinkers are getting the image url from your site directly and not using Google Image search themselves.
So I have painstakingly created another folder and copied all the images into this new folder, leaving the old files in place. Then I re-linked all my images to the pictures in this new folder but I didn't change the file names.
You guys think it'll work?