Research And Implementation Of Two-Dimensional Code Phishing Detection Technology Based On Mobile Platform

Most of the traditional phishing detection solutions are based on computer platforms and are rarely applied directly to mobile devices.With the advent of two-dimensional code technology,the security problem of two-dimensional code phishing attacks has followed,and the phishing attacks faced by mobile phone users have become more serious.In view of the above problems,this paper studies the two-dimensional code phishing detection technology based on mobile platform.The main research contents are as follows:Based on the mobile platform,a method of two-dimensional code phishing detection LogoPhish is proposed.The mobile phone device scans the two-dimensional code image embedded with the logo,and uses the Google search engine to identify the identity represented by the URL(Uniform Resource Locator)in the logo and the two-dimensional code,and then the consistency of the logo and the URL is compared to determine the validity of the two-dimensional code.The experimental results show that LogoPhish has better effect and is superior to traditional detection methods when it is used for the validity detection of two-dimensional codes.Aiming at the problem that the proposed two-dimensional code phishing detection method relies too much on the Google search engine and the logo image size in the two-dimensional code is too small,which leads to the instability of the method,an improved recognition method based on convolutional neural network is proposed.The identification of the logo identity in the two-dimensional code is accomplished by using the improved Faster R-CNN network model.It is verified by experiments that this method is used to identify the logo identity in two-dimensional code with high accuracy.The two-dimensional code phishing detection system based on mobile platform is designed and implemented.The mobile client implements the extraction of the logo in the two-dimensional code and the decoding of the URL information,and the server side implements the logo identification and the consistency matching of the identity.The system test results show that the system can meet the legality detection requirements of the two-dimensional code,and the overall result meets the expected goal.