We are witnessing an exponential growth in the number and size of images in a variety of fields, including: medical imaging, remote sensing, digital libraries, internet, and space exploration. With this growth comes the potential for a wealth of information. However, this can only be possible if we are able to extract from images in raw data format (potentially millions of images, each containing millions of pixels) high-level understanding of their content. The complex nature of images as sources of information makes this task, the task of deriving high-level knowledge from raw data, an increasingly difficult task. Manual interpretation does not provide a long-term and reusable solution. In contrast, maturing techniques in data mining, computer vision, and machine learning, coupled with rapidly improving and affordable parallel and distributed hardware, have the potential to solve current and future image mining problems in an efficient, and scalable manner.

 

In this seminar, we will provide an overview of current research in the area of image mining and knowledge discovery. We will describe our approach in the Pattern Analysis and Machine Intelligence (PAMI) group for developing systems and algorithms for extracting semantically meaningful content from images and text. We will present some cooperative agent-based algorithms that we are developing for discovering knowledge from and about images in the context of the Learning Object Research Network (LORNET).