Security inspection of railway mixed goods is an important guarantee for the safe transportation of mixed goods. To address the problems of a small dataset of dangerous goods, low identification efficiency of manual security inspection, and a wide variety and large quantity of goods, this paper proposed a method for dangerous goods inspection in railway mixed transportation using a convolutional neural network based on data augmentation. Firstly, a generative adversarial network was used to expand the image sample dataset of dangerous goods. Then, the convolutional neural network was used to extract the type features of dangerous goods. Finally, the full sample dataset was trained. With the function loss value, recognition accuracy, recall rate, and harmonic mean of recognition accuracy and recall rate as model evaluation indicators, the experimental results show that the recognition accuracy of dangerous goods by this model reaches 99.5%. The research results can provide technical support for security inspection and dangerous goods inspection in railway mixed transportation.
WANGJiajia, XIONGFeilong. Research on Recognition Method of Heart Sound Signals Based on Improved MFCC and CNN[J]. Computer Measurement & Control,2024,32(12):201-207,215.
GUANYuanlin, LIUGuilin, YUChunyu,et al. Fault Diagnosis Method of Rolling Bearing Based on CNN and BLS[J]. Journal of Vibration,Measurement & Diagnosis,2024,44(5):1017-1022.
WANGHaoxuan, SUShengchao, CUIWenxia. Vehicle CAN Bus Intrusion Detection Algorithm Based on GAN-Transformer[J]. Computer Engineering and Design,2025,46(6):1710-1716.
YUANHoudong, LIHongjiao. Data Repair and Intrusion Detection Method Based on IDM-GAN-APLSTM[J]. Computer Applications and Software,2026,43(2):371-379,397.
HERuimin, CHENGHao, WANGBaiqi,et al. Research on Synthetic-CT Generation from CBCT Using Reg-GAN[J]. Chinese Journal of Engineering,2025,47(9):1853-1861.
[14]
SLIMA, MALAHM. ACGAN:Adaptive Loss-Guided GAN-CNN Framework for Semantically Coherent Synthetic Insider Threat Data Generation[J].Sensors,2024,24(20):6601.
JIAOLiudan, LIUYing, WUYa,et al. Transportation Carbon Emission Prediction Based on Convolutional Neural Network[J]. Railway Transport and Economy,2024,46(8):49-57.
[25]
AL-NAWIN, MADAINA, ABDULLAHM. MishReLU:A Hybrid Activation Function for Deep Neural Networks[J]. Expert Systems with Applications,2026,319:131953.
LIXiang, LEIMiaomiao, LIUGuoxiang. Path Optimization for Road-Rail Intermodal Transport of Dangerous Goods Considering Regional Risk Equity[J]. Railway Transport and Economy,2024,46(11):140-149.
GAOQiang, PANJun, HONGRuifeng. Research on Automatic Recognition of Dangerous Goods in Airport Security Inspection Based on CNN[J]. Computer Technology and Development,2019,29(10):95-99.
[30]
FAH Y, SHUAIB, YANGZ L,et al. Mining the Accident Causes of Railway Dangerous Goods Transportation:A Logistics-DT-TFP Based Approach[J]. Accident Analysis & Prevention,2024,195:107421.
[31]
HUANGW C, SUNL H, YANGZ L,et al. Using Radial Basis Function and back Propagation to Predicate Fault in a Railway Dangerous Goods Transportation System Considering the Markov Correction[J]. Applied Soft Computing,2023,145:110593.