Extraction and Management of Semantic Information from Visual Data

The amount of digital data generated and uploaded in the form of images, on the internet every day by the scientific, medical, educational, industrial and other communities are very large. The difficulty of retrieving the most similar images from a database of images increases with the number of images in the database. On the other hand, the user queries are becoming very subjective and traditional text-based methods cannot efficiently handle them. The subjectivity of human perception and the rich contents of the images further aggravate the problem. To overcome this, a new query-by-example technique using multiple colors, texture and shape features is proposed in this paper for the extraction and management of semantic information from visual data. The experimental results suggest that our proposed technique is efficient and retrieves semantically more similar images.