Fishing webpage detection method based on image processing

Abstract

The invention relates to the field of image information processing and network security, in particular to a method for detecting a phishing webpage based on image segmentation. The invention calculates the similarity between the suspicious phishing webpage and each protected webpage one by one, and determines that the suspicious webpage is the phishing webpage as for the protected webpage, when the similarity thereof with the protected webpage is smaller than the preset value. The invention is characterized in that the webpage image visited by users can be directly analyzed, with high reliability and difficulty to be cheated by webpage design technique; by segmenting a webpage picture, partial matching can be realized, preventing network phishing caused by partial similarity; the matching method that pays equal attention on sub-image properties and relationship meets visual psychology better and makes the matching results more accurate. According to the data obtained by experiments, the invention has the advantage of high detection precision of webpage image.

Description

technical field [0001] The invention relates to the fields of image information processing and network security, relates to an anti-phishing method, in particular to a method for detecting phishing webpages based on image processing. Background technique [0002] Phishing (Web Phishing) is a kind of online identity impersonation, which uses social engineering (that is, deception) and technical means (including malicious code, phishing emails, monitoring software, etc.) to steal personal identity information or account numbers of network users, Passwords, credit card numbers, etc. Phishing has now become a great threat to user information security, and has caused huge economic losses to users (according to authoritative organizations, the loss caused by phishing in the United States in 2007 reached 2.8 billion U.S. dollars); It has seriously hindered the development of electronic finance such as online banking and online payment in my country. [0003] At present, the common...

AI Technical Summary

Problems solved by technology

[0007] However, the above three methods all have great deficiencies. The first method cannot detect emails containing images, including links provided by using pictures and mapping methods or addresses displayed in pictures, etc.; the second method will greatly reduce the network speed. It is also extremely complicated to maintain and involves user privacy, so it is difficult to deploy; method 3 cannot deal with bridging attacks, and a simple automatic submission prevention mechanism (verification code) can make this party invalid

Images