| 摘要 |
In an image obtained by reading a document or the like, when a round portion is present at a corner of a frame for describing a character, extraction of a region where a character image is described is performed by removing pixels constituting a part of rules or lines constituting the frame. Such pixels are removed from an extracted image region by identifying whether or not a desired pixel group contacting at least one of row-direction rules and column-direction rules constitutes part of the rules, by performing labeling for the pixel group.
|
