Registration and Fusion Using Reflection in Passive Millimeter-Wave Images

Abstract

Passive millimeter-wave (PMMW) imaging technology is widely used in civilian and military applications. However, there are reflections similar to the optical band in PMMW images, which have a negative influence on target detection and recognition. In this article, we present a reflection-based method to enhance the target features in PMMW images. The dividing line between the target and reflection is obtained by the similarity of brightness temperature (TB). By combining the similarity and reflection principle, we propose a new method to obtain the feature points of the target and reflection for registration. Then, the weighted method based on region TB is used to fusion target and reflection. Finally, to avoid interference with target detection and recognition, the reflection is removed. The experimental results show that the method can obtain higher contrast and more accurate target information.