Patent №
US 9,501,696
Granted
—
Owner
—
Lab
—
AI components
5
ml · nlp · vision · kr · hardware
Assignment
None on record
Dataset
AIPD
2023_r1 edition
Application
15019164
The present disclosure provides a system and method for mapping zone-based metadata into one or more reference-able objects. The system and method extracts and/or generates metadata from a document according to custom zones defined in a document type. Each custom zone is associated to actions such as generating events, invoking triggers and/or invoking callback methods. One or more of these actions may be triggered when pre-processing or post-processing a document. According to an inventive concept of the present application, when a document is stored in a data storage and placed in a process queue to be processed by a suitable processing plugin according to the custom zones defined in the document. For instance, an image file comprising textual custom zones may be processed by an OCR plugin which extracts text metadata from the document. Other processing plugins include plugins to extract metadata from proprietary document formats such as CAD, PPTX, PDF, PSD, or others. Plugins are also available for generation of metadata by interpreting an image (i.e. detection or identification of objects and/or scenery in an image) through artificial intelligence and computer vision means. The process queue can be used for scalability of the quantity of actions to process.