This script queries images.google.com for each keyword given. In the big text box you can put some text where the urls will be harvested from. For each url in this textbox google will be queried for each keyword, so google will be asked if each keyword occurs on each url (in google you would have to enter for each keyword and each url: 'keyword site:url').

This tool queries sets of 20 till the max nr of results of google or the max nr of user defined results is reached. If all scrapes are done a list with urls for all the images is displayed

Warning: google uses it's own algorithms for determing if a certain image belongs to a certain keyword. This is by no means exhaustive, correct or complete. An other way to get labels for images in a certain page is through checking the html 'alt'-attributes of an image. You can try this with the Image Scraper.
Topic revision: r1 - 19 Dec 2008, ErikBorra
This site is powered by FoswikiCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding Foswiki? Send feedback