This article focuses on research into the use of content-based image retrieval in web searches, instead of keywords. Typing into Google Image the single word "rosebud" returns about 60,000 pictures. But it is unable to differentiate between the flower in bloom and legendary film director Orson Welles's scowl. Ideally, an Internet user should be able to use the likeness of a rose to tell a search engine to find others like it. Major search engines have yet to implement this form of image retrieval to mill through their indexes of images. Still, research by both industry and academia has achieved some intriguing advances of late that sidestep the need for keywords--and address the challenge of analyzing the content of images in large databases.
© 2001-2024 Fundación Dialnet · Todos los derechos reservados