Skip to Main Content
Image spam is a recent variant of spam, which can easily circumvent widely text spam filters. In this paper, we propose a new method of filtering image spam, which is called FH (file properties and histogram) algorithm. FH algorithm utilizes file properties and histogram (gray histogram or color histogram), and is a fast method because of not extracting text and analyzing the content of Email. A preliminary experimental evaluation of our approach is reported on a personal dataset and Princeton benchmark dataset of spam images publicly available.
Date of Conference: 30-31 Dec. 2008