Annotation Data Generation and Overlay for Enhancing Readability on Electronic Book Image Stream Service

摘要

Provided are techniques for receiving a scanned image corresponding to a page; analyzing the scanned image to identify a set of characters (Ci) and corresponding positions (Pi) of each character of Ci on the page; applying natural language processing (NPL) and an analytic analysis algorithm to determine a semantic analysis relationship of phrases formed by Ci to determine meanings for the phrases; generating a plurality of annotations (Ai) indicating the determined meanings of the phrases, wherein each annotation is assigned a corresponding position (Qi) on the page based upon the Pi of the corresponding characters on the page; and storing, in a non-transitory computer-readable medium, the scanned image in conjunction with the plurality of annotations and the corresponding Qi.

主权项

1. A method, comprising:
receiving a scanned image corresponding to a page of text; analyzing the scanned image to identify a set of characters (Ci) and corresponding positions (Pi) of each character of Ci on the page; applying natural language processing (NPL) and an analytic analysis algorithm to determine a semantic analysis relationship of phrases formed within Ci and to determine meanings for the phrases; generating a plurality of annotations (Ai) indicating the determined meanings of the corresponding phrases, wherein each annotation of Ai is assigned a corresponding position Qi on the page based upon the Pi of the corresponding characters on the page; and storing, in a non-transitory computer-readable medium, the scanned image in conjunction with the plurality of annotations and the corresponding Qi.