Research On Synthesizing Method Of Security Inspection X-ray Image Based On Deep Learning

Using deep learning to assist people in recognizing prohibited items in X-Ray images is crucial to improve the quality of security inspections.However,these methods based on Convolutional Neural Network(CNN)require lots of data,and the data collection usually takes a lot of time and effort.In this paper,we propose a method to synthesize security X-ray images based on Generative Adversarial Network(GAN).The main contributions are as follows:1)A synthetic model of security X-ray image is established.First,the proposed framework is built on the Generative Adversarial Networks(GAN)with multiple discriminators,trying to synthesize realistic X-Ray images,and improving it from three aspects: network structure,loss function,and parameter settings.2)A guided filter is introduced in the GAN synthesis model.It is used to solve the problem of lack of detail and texture in model security X-ray image synthesis.Then,the Frechet Inception Distance(FID)evaluation index is used to quantitatively analyze the performance of different GAN models.3)Data enhancement verification is performed on the synthesized image based on the SSD network.First of all,we use the real image data set as the training set to train the SSD model,and then add synthetic images to the training set to train the SSD model.The experimental results show that the images synthesized by XS-GAN model can effectively achieve the effect of data enhancement.