Spider's Latest

Google trumpets new image search system

Google has revealed that it has developed new technologies which could make it easier for people to search for images.

The new tool will use an algorithm similar to PageRank which will be able to assess the similarity of images and provide better results for users.

In a research paper titled PageRank for Product Image Search, Google analysts Shumeet Baluja and Yushi Jing explained that the new system had already shown promising results in tests.

They found that people using the new system found 83 per cent less irrelevant images in searches.

The researchers explained how the system worked: "By treating images as web documents and their similarities as probabilistic visual hyperlinks, we estimate the likelihood of images visited by a user traversing through these visual-hyperlinks.

"Those with more estimated visits will be ranked higher than others."

Andy Beal explained on the Marketing Pilgrim blog that Google was pursuing "the Holy Grail of image search" and said that the system would only be successful if it could be "applied to the billions of images floating around the web". ADNFCR-1536-ID-18571688-ADNFCR