A X-ray Security Inspection Contraband Detection Method Aiming at the Overlapping Problem

Abstract

Aiming at the overlapping problem of X-ray images existing in the detection of contraband in X-ray security inspection, the present invention proposes a detection method for contraband in X-ray security inspection, which can be deployed on a security inspection machine for use. The present invention is realized through the following technical proposals: step 1, input training set images into the network for training; step 2, input test set images into the trained network for testing. Compared with the prior art, the present invention has the following advantages: 1. The present invention adopts an end-to-end deep learning network, which has fast detection speed and is easy to deploy. 2. The present invention only needs to use the images generated by the X-ray security inspection system to make predictions. Compared with other methods, it does not require additional physical equipment. The method is simple and easy, and can be quickly deployed on security inspection machines and put into practical applications .

Description

technical field [0001] The present invention relates to the field of computer vision. Background technique [0002] With the increase of crowds in public transport hubs, security checks are becoming more and more important in protecting public safety. X-ray baggage security inspection systems are usually operated by human inspectors, but the method of simply relying on human eyes to search for dangerous goods in X-ray images has great instability. When the dangerous goods in the sealed package present an unrecognizable view in the X-ray image due to occlusion and rotation, it will make the detection task very difficult. Especially in the case of dealing with a large number of security inspection pictures, the staff only have a few seconds to judge whether there is a threat in the bag. In the inspection process of a large number of objects of various shapes and materials that do not contain threats, it is difficult to maintain continuous concentration. Misjudgments and miss...

AI Technical Summary

Problems solved by technology

However, the whole method needs to be based on a large number of physical conditions, such as nuclear density meters, spectral analyzers, etc., and the training process is complicated. When the category increases, the entire classifier needs to be retrained and the network modified. This method of individually training classifiers for each category is used in practice. difficult to achieve in the process

[0006] In addition to the above-mentioned problems, the more important problem faced by the detection of contraband in X-ray security inspection is the overlapping problem of X-ray images, and the existing technology does not deal with this problem

Images