Image processing method and apparatus which orders text areas which have been extracted from an image
AssigneeCanon Kabushiki Kaisha
InventorTadanori Nakatsuka
Life of this patent
Estimating the term…
Abstract
An apparatus and method for automatically determining the order of reading a plurality of text areas, when an original constituted of the plurality of text areas is input as an image. For this purpose, the structure of text areas is analyzed hierarchically by dividing text areas into groups (plural text areas constituting one article) in consideration of the positional relation of the text areas with a separator, a figure and text areas, rather than simply ordering from top to bottom, so as to also cope with a complex layout such as a newspaper. Also, the connectivity of text areas is judged by making use of information, such as the text contents of the text areas or a connection mark in the text.
Claims14
An image processing method, including the steps of:
- inputting an image;
- extracting text areas from the input image;
- dividing the text areas into groups;
- and
- determining the ordering of the text areas within the groups.
An image processing apparatus comprising:
- image input means for inputting an image;
- text area extraction means for extracting text areas from the input image;
- grouping means for grouping the text areas;
- and
- order determining means for determining the ordering of text areas within the groups.
Dependencies are inferred from the claim text.