Share Email Print

Optical Engineering

Integrated and hierarchical sortmap-relabel image segmentation methods
Author(s): Lei Ma; Jennie Si; Glen P. Abousleman
Format Member Price Non-Member Price
PDF $20.00 $25.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

We propose two image segmentation algorithms: the integrated sortmap relabel with adjacent-region merging (ISARM), and the self-guided sortmap relabel with adjacent-region merging (SGSARM). Due to the integration of noise reduction and fast merging, ISARM provides a 25% improvement in processing time, as compared to leading existing algorithms such as the region adjacency graph (RAG) algorithm, on a variety of test images. ISARM also provides better segmentation accuracy than the RAG algorithm, by a measure combining the mean squared error and the number of regions obtained. SGSARM is designed for use with large images (say 1024×1024 or larger). It incorporates two levels of processing: an edge detection algorithm of linear complexity, which is applied to large images to detect regions of interest (ROIs), followed by ISARM for finer segmentation of each ROI. SGSARM therefore has significant advantages in speed and accuracy when used in large images. Simulation results are provided to demonstrate the performance of both algorithms.

Paper Details

Date Published: 1 November 2002
PDF: 10 pages
Opt. Eng. 41(11) doi: 10.1117/1.1511246
Published in: Optical Engineering Volume 41, Issue 11
Show Author Affiliations
Lei Ma, Arizona State Univ. (United States)
Jennie Si, Arizona State Univ. (United States)
Glen P. Abousleman, General Dynamics Decision Systems (United States)

© SPIE. Terms of Use
Back to Top