Hierarchical partitioning and contour extraction algorithm
Conference Paper
Overview
Additional Document Info
View All
Overview
abstract
The problems of deriving hierarchical regions from images and their corresponding contours are considered. The proposed hierarchical partitioning method produces the region specifics: the median, lowest and highest values, size, and the locations of the pixels for each region. Hierarchical partitioning is achieved using the 'pixel value proximity of intensity' criterion. The partitioning algorithm employs a realtime algorithm that computes the region specifics. The proposed contour extraction method repeatedly selects an optimal pixel among many neighbor pixels until no further eligible pixel is found on the given contour. The hierarchical partitioning and contour extraction algorithms have been implemented and their results are discussed.