|
Activity Number:
|
215
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 3, 2009 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Statistical Learning and Data Mining
|
| Abstract - #304841 |
|
Title:
|
Image Segmentation by SUP Clustering Algorithm
|
|
Author(s):
|
Ting-Li Chen*+
|
|
Companies:
|
Institute of Statistical Science, Academia Sinica
|
|
Address:
|
, , , Taiwan
|
|
Keywords:
|
image segmentation ; image processing ; clustering
|
|
Abstract:
|
Image segmentation plays a very important role in image processing. There are various approaches in this field. One of them is by clustering algorithm. Each pixel in an image can be treated as an individual subject, and the goal of image segmentation is to assign these subjects into clusters with similar pixels grouped in the same cluster. If only intensity information is considered, clustering results are usually poor, especially for noisy data. The location information is another useful variable for segmentation. However, most popular clustering algorithms do not perform well even after the location variable is taken into consideration. We applied SUP clustering algorithm (Chen & Shiu 2007) on the segmentation problem and obtained good results. We will present how to choose parameters in SUP algorithm for image segmentation and how to combine information from intensity and location.
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2009 program |
