Search Space Partitioning Using Convex Hull and Concavity Features for Fast Medical Image Retrieval

A new approach is presented for partitioning an image database by classifying and indexing the convex hull shapes and the number of region concavities. The result is a significant increase in image search and retrieval speed. The convex hull is first determined using a novel and efficient approach based on the geometrical heat differential equation. A set of seven templates is available to approximate the boundary as observed from a particular viewpoint. The convex hull is then represented by a vector of three such shape template indices as observed from three viewpoints. This result enables the regions in the image database to be divided into 344 convex hull classes. Concavity information, obtained using a boundary support parameterization, further partitions the database. A given query must now be compared only to shapes of a few classes, hence searching is much faster. Both theoretical background and practical results are discussed.

• Type: Conference
• Year: 2004
• Where: ISBI
• Authors: Phillip Mlsna, Nikolay Sirakov
• Keywords: Convex Hull, Convex Hull Classes, Convex Hull Shapes, ISBI 2004, Medical Imaging