Newspaper Segmentation and Adopted Methods

This technical paper deals with the segmentation of newspaper articles into the distinct regions of text (i.e. paragraphs), pictures and background. It has been written to gather and consolidate ideas for personal use. I''ll basically outline the aim in more detail, along with assumptions and methods of segmentation. The methods will be presented as generic methods, current hacks and standard methods which may be useful. Some material presented may have little to do with newspaper segmentation, but has been included for future reference and to organise ideas.