Detecting Content Based Image Spam in E-mail

In modern Technology, E-mail is becoming one of the fastest communication because it is very cheap and easy to send, one of the measure concern in email is security there is lots of fraud happening through spam emails. There are various techniques to detect image based spam emails such as image spam filtering tool which is able to detect image spams but they have some limitation. In this paper we proposed a method which can detect an image based spam. In this method we describe into two parts first we receive an image based spam mail from receiver side then extract characters from those images using OCR (optical character recognition) techniques, the characters are compared with the database of words suspected to be used by spammers and secondly the remaining parts of the images that can compare into the proposed database. The important of this paper is we can detect pure images.