An Two-dimensional Barcode Security Technology Based On Time-domain And Frequency-domain Features

With the rapid development of digital technology,two-dimensional code can be seen everywhere in our life.Two-dimensional(2D)barcode technology is widely used in various fields,such as commodity security,mobile e-commerce.However,due to the simple structure and low difficulty of production,it is easily forged by criminals.In the field of commodity safety,the 2D code is affixed as an anti-counterfeit label on the surface of the product packaging,and the criminals can easily copy it and print it on counterfeit goods.In addition,the means of criminals are diverse,the development of anti-counterfeiting technology still has enormous challenges.In this context,we aim at proposing an authentication scheme with a mobile imaging device focused on a given 2D barcode.The main research work is as follows:(1)Two channel models related to 2D barcode generation process are established:Print-Capture channel(PC)and Print-Scan Print-Capture channel(PSPC).Based on these two channels,the proposed scheme of 2D barcode authentication is mainly used to analyze the structural similarities and differences between 2D barcode images in frequency domain and time domain,so as to achieve the purpose of authentication;(2)First,we designed a gray-scale 2D barcode that uses the characteristics of the print-capture/scan channel model.Different from the conventional 2D barcode,the gray part of the 2D barcode is processed by the halftone technology to form a regular array area,which is the core part of this paper;(3)The dot matrix in the above-mentioned 2D barcode can be regarded as an impulse sequence in a 2D plane.By analyzing the position and number of these impulse points,the frequency domain features of the 2D barcode are constructed;in time domain,an improved LBP descriptors which accords with the above mentioned gray-scale 2D barcode is designed based on standard LBP descriptors.Then,by combining the two features in a cascade manner,an efficient two-level barcode authentication framework is formed;(4)To evaluate the practicality of the proposed authentication scheme,five databases with different devices(printers,scanners,mobile cameras),barcode sizes,barcode designsand counterfeit barcode generated by more sophisticated attacks are considered in the experiments.The number of 2D barcode in all databases can meet the experimental requirements of this paper.By comparing with the existing texture descriptors and some deep learning-based approaches,it is shown that the proposed scheme has a higher authentication accuracy under various conditions,such as cross-database,cross-size cross-pattern and dealing with advanced means of attack experiments which study the generalities of a pre-trained model towards challenging conditions commonly found in real-world scenarios;(5)In this paper,we also do an expandable experiment to study the security of documents.Taking the certificate as an example,we modified the frequency domain features based on the idea of 2D barcode authentication scheme by extracting the image frequency domain information on the scanned certificate to make it conform to the features of the scanned certificate image.Finally,the experiments result show that the features we extract in the frequency domain are also applicable to document security.