Improved Threshold-based Segmentation Method for Millimeter Wave Radiometric Image

DOI

10.2991/smont-19.2019.38How to use a DOI?

Keywords

millimeter wave; image segmentation; method improvement

Abstract

Working all-day and all-weather, the passive millimeter wave radiometer can be used in many fields to detect objects, especially for the concealed objects. With the passive millimeter wave radiometric image processed and analyzed, we can get the shape or the center of the interested object which may be helpful for the following operation. However, some classical segmentation methods can’t work well for the passive millimeter wave radiometric image with the existence of transition band near the edge of object. Therefore, we propose a simple improved segmentation method based on the maximum between-class variance and the maximum entropy threshold selection method. To be specific, when finding the right threshold for object segmentation, the difference between the first and second segmentation result based on the maximum between-class variance threshold selection method is first obtained so we get the approximate location of the edge of the target. Then we can get the final threshold for segmentation by applying the maximum entropy method to the obtained local region. The improved method takes not only the advantage of the two threshold selection methods, but also fully considers the global and local information. Experimental result shows that the method has a better segmentation effect for our passive millimeter wave radiometric image and the calculation is relatively less.

Copyright

© 2019, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Changchang Yu
AU  - Guangfeng Zhang
AU  - Yuan Gao
PY  - 2019/04
DA  - 2019/04
TI  - Improved Threshold-based Segmentation Method for Millimeter Wave Radiometric Image
BT  - Proceedings of the 2019 International Conference on Modeling, Simulation, Optimization and Numerical Techniques (SMONT 2019)
PB  - Atlantis Press
SP  - 169
EP  - 171
SN  - 1951-6851
UR  - https://doi.org/10.2991/smont-19.2019.38
DO  - 10.2991/smont-19.2019.38
ID  - Yu2019/04
ER  -