GazoPa is a web image search service that uses features from an image to search for and identify similar images. In conventional image search, users do not find results of image searches to be as accurate as those of web page searches. This is often because some images do not have metadata, some have incorrect metadata, and some are difficult to describe with words. Therefore, keywords are not sufficient as the only conditions for image searches.
With GazoPa, users can overcome the limitations of metadata, and word descriptions. GazoPa enables users to search for similar images using characteristics such as a color or a shape extracted from the image itself. GazoPa even enables the use of users' own photos, drawings, and images found on the web, as search keys to locate similar images from the GazoPa database. Not only does GazoPa support photos but also searches video thumbnails. Unlike video sharing websites that use keywords to search, users can search for videos using images.
GazoPa enables searches at high speed even for large quantities of image data. GazoPa currently searches 50 million images crawled from the web within one second. Since GazoPa crawls the web continuously, it will soon be capable of searching more than 100 million images.
As the number of digital camera and camera phones increases steadily to exceed 1 billion worldwide, the number of digital images that are captured by digital devices also increases dramatically. It therefore becomes more and more important to search for a needed image from a large quantity of images in a short time. GazoPa overcomes the limits of keyword searches and introduces a new world to image searches.-Hitachi
