Content-based image retrieval

Content-based image retrieval (CBIR) refers to techniques used to search for digital images by features of their content, which is particularly helpful when studying large databases. It is often preferable to perform searches relying on metadata, which can be expensive and time-consuming to produce, as it requires humans to describe each individual item in the database.

Searches can be performed by providing an example image, e.g. a pre-existing picture, or a drawing by the user, which the system then uses as a basis for its search. This process is called ‘query by example’. ‘Semantic retrieval’, i.e. searching using keywords, is still quite difficult for computers to perform.

The search analyses the actual contents of the image, e.g. colour, shape and texture, using techniques from statistics, pattern recognition, signal processing and computer vision.

Some systems use ‘relevance feedback’, where the user rates the relevance of each search result. This can help to refine the process.

CBIR can also be termed ‘Query By Image Content’ (QBIC) and ‘Content-Based Visual Information Retrieval’ (CBVIR).

Related methods include: Content analysis, Content-based sound retrieval, Searching and querying and Statistical analysis.

Syndicate content