Digitized images are associated with English language captions and other data, collectively known as the metadata associated with the images. A natural language processing database removes ambiguities from the metadata, and the images and the metadata are stored in databases. A user formulates a search query, and natural language processing is used to determine matches between the query and the stored metadata. Images corresponding to the matches are then viewed, and desired images are selected for licensing. The license terms for selected images are displayed, and a subset of the selected images are ordered as desired by the user.