Research on Contraband X-ray Image Recognition Method Based on Convolutional Neural Network

With the expansion of the scope of human activities and the increase of mobility, public security is facing more and more severe challenges. The traditional manual inspection method can no longer meet the new challenges faced by the current security inspection. For the new security inspection requirements that take into account both accuracy and processing speed, this paper uses YOLOX-Nano to build the original target detection model and improve it. In order to improve the receptive field of the backbone network and retain sufficient shallow feature information, a multi-branch feature extraction module is introduced at the bottom of the model. At the end of the model, the Convolutional Block Attention Module (CBAM) attention mechanism is introduced to improve the feature extraction ability of the network, and the Swish activation function is used as a new activation function to make the network achieve better accuracy and generalization. The experimental results show that the improved model improves the mean average precision (mAP) and COCO mAP by 0.34% and 1.3% respectively.