Image Fusion Based Concealed Weapon Detection

Abstract

Two concealed weapon detection methods based on image fusion are presented. The dual tree complex wavelet transform (DT-CWT) is used to image fusion processing because of its shift invariance and directional selectivity, and it is preferable to traditional discrete wavelet transform (DWT). The fusion rules based on multi-scheme are put forward in the gray visual image and millimeter wave image fusion. The DT-CWT is combined with color space to fuse color visual image and infrared image. Two experiments indicate that the concealed weapons are detected and the detail information of visible images is expressed clearly. And it also shows that DT-CWT is superior to DWT.