There are estimated to be over 370 million pornographic pages on the internet and thousands more being created every day. (Should you repeat content – also in web categoriser?) In addition to this the new concern is that pornographic images can easily be access via image search engines and on social networking sites. A core question then is, how can parents protect their children against this uncensored blight which fills the web? The issue is a worldwide problem effecting internet users in every country of the world. The dynamic nature of the internet makes it difficult to control, as pornographic domains are changing every day.

Consumer Web Filtering products often contain URL blacklists and real time analysis features. Such heuristic filtering technologies are normally based purely on lexical analysis, which can be ineffective on pornographic sites with minimal or no text content and foreign language sites. There is however, one constant that exists across all pornographic website regardless of language or country of origin – illicit images.

When performing real time inline web filtering the best practice is to employ a layered approach utilising multiple analysis methodologies which are layer based on performance.

For consumer web filtering products to effectively filter pornographic content it is essential to analyse the images on the page in conjunction with the other elements such as the domain name and text.