An Information-Driven Framework for Image Mining

Image mining systems that can automatically extract semantically meaningful information (knowledge) from image data are increasingly in demand. The fundamental challenge in image mining is to determine how low-level, pixel representation contained in a raw image or image sequence can be processed to identify high-level spatial objects and relationships. To meet this challenge, we propose an efficient information-driven framework for image mining. We distinguish four levels of information: the Pixel Level, the Object Level, the Semantic Concept Level, and the Pattern and Knowledge Level. High-dimensional indexing schemes and retrieval techniques are also included in the framework to support the flow of information among the levels. We believe this framework represents the first step towards capturing the different levels of information present in image data and addressing the issues and challenges of discovering useful patterns/knowledge from each level.

• Type: Conference
• Year: 2001
• Where: DEXA
• Authors: Ji Zhang, Wynne Hsu, Mong-Li Lee
• Keywords: Database, DEXA 2001, Image Data, Image Mining, Image Mining Systems