Enhancing backscatter images with transmission images in security inspection system
You Xiaoming, Chang Qingqing · 2012
We propose a novel scheme for enhancing backscatter image by fusing it with simultaneously captured transmission image in security inspection system. The backscatter image is sensitive to organic matter, such as drugs and explosives. However, it is noisy and lacks fine structural information which can be provided by the transmission image. Our enhancing method is constructed by three parts: image preprocessing, image filtering and image fusion. In the preprocessing step, intensity calibration and image registration are performed on the backscatter image. Then one new method called dual bilateral filter is introduced, which is applied for texture extraction and backscatter image denoising. Finally, the dual filtered backscatter image and the extracted textures are fused together to obtain one new image with more details and better perceptual quality. Experimental results show that the proposed method has good performance for noise reduction and backscatter image enhancement.