The Design And Realization Of Image Spam Interception System

With the rapid development of information technology, e-mail has replaced the traditional letters. Then the spam appears and makes e-mail be filled with all kinds of invalid information. It makes people can’t concentrate on care content, but struggle to cope with something that is nothing to do with them, even ignore the important messages.In recent years, spam is developing in diversifications, including more multimedia content, such as images. Image is a kind of multimedia information, it contains a large amount of bytes and its content of the description is fuzzy. So searching a specific content in the image is much more difficult than text. In conclusion, it is necessary to construct a new interception system to cope with the development of the traditional interception system.In order to effectively block the spam mails, this paper proposes an image matching algorithm based on corner features. This feature has some advantages. Its algorithm is simple, its matching speed is fast, and it is not affected by external factors such as brightness. Harris algorithm can be used to quickly detect the corners of the image. Then the corner feature is extracted from the neighborhood of the corner and feature database is established by using these features. When an image is input into the system, its corner features are also be extracted. Then similarities between its corner features and the corner features in the database are compared. If similar number of corner points is large, the image is judged as an image spam and intercepted. The proposed algorithm can locate the position of similar corners from the image, and further determine the similarity of the image, which is suitable for a large number of image matching tasks.A spam mail interception system is designed and implemented in this paper. The system is developed by Visual Studio, Server SQL and Jmail component. Server SQL is used to establish a database and Jmail is used to send and receive e-mails. The system can effectively intercept similar images in the blacklist database, and is not a greater burden of the interception system.