Dual-energy X-ray Image Segmentation And Identification Technology And Its Applications

In order to fight with terror activities with bombs which are bringing more and more harm to the human society, bomb detecting devices have been widely used in airports, railway stations, and other important places. However the current detection devices cannot work fully automatically, the final recognition procedure must be done by human operator. Therefore a lot of manpower and money have been put into the fight, to enhance the security check by manual work on one hand, and to develop more effective detecting techniques on the other hand.The main goal of this paper is to develop digital image processing and pattern recognition methods to detect hidden bombs automatically. With the images gotten from widely used dual-energy X-ray system, this paper puts emphasis on the research of image segmentation in value domain and space domain, and bomb image recognition by computer programming techniques. In the research of value domain segmentation of images, a modified variance algorithm, a subtraction segmentation algorithm and an improved pulse coupled neural networks segmentation algorithm are proposed. In order to successfully implement the space domain segmentation, this paper also develops an algorithm that can distinguish connexity between pixels and extract the feature parameters in one image scanning procedure. Based on the proposed algorithms a decision rule is given for bomb target identification of dual-energy X-ray images. Simulation experiments have been made. And the results show that, the correct recognition rate with above 90% is achieved by using the algorithms given in this paper.