| 摘要 |
<p>Characteristics of images, such as skew of lines of text or dominant typeface of characters, are detected by producing distance data for each of a number of starting pixels within an image. Each starting pixel meets a criterion, such as an edge pixel or a pixel inside a connected component. Each starting pixel's distance data indicates the distance in each of a number of directions from the starting pixel to another pixel at which the image meets another criterion. For example, to detect skew of lines of text, the distance data can indicate distance from a starting pixel at an edge across white pixels to another edge. To detect dominant typeface, the distance data can indicate distance from a starting pixel at an edge or inside a connected component across black pixels to another edge. The separating angle between adjacent directions can be small enough to detect features of an appropriate size at an appropriate distance, such as features of character size at the average intercharacter spacing. The distances at each direction from all the starting pixels can be averaged to obtain a combined distance at each direction. A sufficient number of starting pixels can be used so that the combined distance is approximately the same as if every pixel in the image were a starting pixel. The combined distances at all the directions form a profile. Skew can be determined from peaks in a profile. A profile can be compared with each of a set of model profiles to detect typeface. <IMAGE></p> |
