TinEye was launched in open betaIdee has just launched the open beta version of its latest product, an image search engine that promises to revolutionize the market. TinEye is now available to all those who want to see whether their images uploaded on different websites have been used illegally, or who just want to check if their work was useful to other people, in case we're talking of a free image.
The new search engine stands apart from its predecessors or competitive products by the fact that it doesn't recognize only perfect matches, but also takes into account cropped and resized images. Also, TinEye identifies an image even if it has been edited with different software applications.
Users have two standard possibilities of searching for the "lookalikes" of an image: they can either upload the image and then search for matches, or just upload a link that corresponds to the location of that particular image. The second the upload is made, Idee's product searches through more than 487 million pages for a match. This number of pages is the maximum that the team has managed to index until now, but TinEye is still in its beta phase, so they hope to have, soon enough, billions of webpages to search through.
"Unlike any other search engine, when TinEye finds a picture, it looks at the pixels in the image, using Idee's identification technology. A unique digital fingerprint is made for each image found." says an official of the company.
When someone performs a search, this fingerprint is compared with every single fingerprint in the indexed pages. The differences between the original image and a copy on the Internet can be easily seen, as users also have the possibility of clicking the thumbnails of the displayed results. Also, the images that resemble the original to the greatest extent are displayed on the first page, while those that have been severely modified appear on the following pages.
TinEye also offers a great tool for Firefox users. A plug-in displays for them a TinEye menu every time they right-click on any given image on the web. This makes image search even more reliable and easier to handle.