Automatic Recognition Of Prohibited Items Based On Multi-energy X-ray Imaging

With the continuous deepening of global economic integration,international economic and cultural exchanges have become more frequent,and the inspection and quarantine work based on X-ray security inspection has been on the increase.At present,the security inspection of quarantine mainly relies on security staff.The use of automatic detection and identification of prohibited items based on security images,can reduce the workload of staff,and prevent high quarantine risk items to enter our territory economically and effectively.Based on the multi-energy X-ray image,we extracted the features of prohibited items and used neural network to recognize the target objects automatically,preliminarily established a set of automatic identification scheme aimed at high risk of prohibited items.At the same time,in view of the problem of overlapping between objects in the traditional X-ray image,a new type of multi-angle X-ray security equipment model and its reconstruction algorithm are proposed,which can use the incomplete X-ray image from multiple perspectives to reconstruct three-dimensional space information of the detected object.The simulation experiments showed that the multi-angle X-ray security inspection model and the reconstruction algorithm proposed in this paper could effectively improve the overlapping between objects in the security inspection,and ensure the image quality and improve the automatic recognition rate.The innovation of this paper mainly focuses on the research of automatic detection and identification of high-risk quarantine items.At the same time,a new multi-angle X-ray security inspection model and its reconstruction algorithm are proposed,and it's effectiveness of the problem overlapping between objects in the security inspection has been verified.