Detecting continuation of flow of a page

摘要

A flow of objects, such as text, on a page may be analyzed by generating an initial determination of whether a portion of an object is continued from one page to another based on a cue in an area of the object. The initial determination may be visually represented on a page. The initial determination may be corrected by an editor to create an editor determination when the initial determination is not a correct continuation status of the portion of the object. In some aspects, the initial determination may include a statistical confidence interval, a reason display, and/or a magnification of portions of the object which may be included in a continuation. In other aspects, heuristics may learn from the editor changes to improve a subsequent initial determination.

主权项

1. A computer-implemented method comprising:
as implemented by one or more computing devices configured with specific executable instructions, generating a first determination of a first continuation status of text in a first text group on a first page of a document that is separated from a second text group on a second page of the document, the first determination being based at least in part on a contextual cue within the document associated with at least one of the first text group or the second text group, wherein the first continuation status indicates that the second text group is a continuation of the first text group; visually representing the first determination to indicate the first continuation status; receiving an indication to confirm the first determination; generating continuation data, based at least in part on the received indication, wherein the continuation data indicates that the first text group and the second text group are connected; generating a reflowable image-based file including the first text group, the second text group, and the continuation data; and applying the continuation data from the reflowable image-based file to the first text group and the second text group from the reflowable image-based file to cause presentation of the reflowable image-based file with the first text group and the second text group presented as connected.