|
Title:
|
A PLATFORM FOR REGION SPACE ANALYSIS IN BINARY PARTITION TREES |
|
Author(s):
|
Huihai Lu , John C. Woods , Mohammed Ghanbari |
|
ISBN:
|
978-972-8924-30-0 |
|
Editors:
|
Nuno Guimarães and Pedro Isaías |
|
Year:
|
2007 |
|
Edition:
|
Single |
|
Keywords:
|
Object extraction, image segmentation, image filtering |
|
Type:
|
Full Paper |
|
First Page:
|
197 |
|
Last Page:
|
204 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
click to dowload 
|
|
Paper Abstract:
|
In this paper, we present a region-based image analysis platform. Within a human in the loop, it provides an efficient method for object extraction, image segmentation and filtering. The region space is organised in the Binary Partition trees which are created based on merging initial small homogenous regions until a single region representing the entire image is obtained. The hindsight of the merging process is therefore stored within the tree structure and region evolutions from seed regions to the root can be derived. By analysing these evolutions, discontinuities caused by merging statistically disjoined regions are identified and the corresponding tree nodes are collected, which highlight the salient image details. Based the detected nodes, tree simplification and image filtering are derived. Given a simpler version of the original tree, manually assisted object extraction and image segmentation can be easily achieved and the experimental results show that our method highlights semantic content in the image. |
|
|
|
|
|
|
