Method for extracting and displaying the full content of selected drawing shapes

HU231764B1Active Publication Date: 2026-03-28MISKOLCI EGYETEM
2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
HU · HU
Patent Type
Patents
Current Assignee / Owner
MISKOLCI EGYETEM
Filing Date
2018-11-28
Publication Date
2026-03-28
Patent Text Reader

Abstract

The subject matter of the application is a method for extracting and displaying the full content of selected drawing shapes during a computer process, according to which, after document input, data that can be displayed in digital form is extracted using an optical character extraction method, then displayed and / or stored in such a way that, after document input, shape selection is performed using the shape management method and then sample loading or shape selection is performed using an operator, then character extraction is performed using the method in the thus selected open or closed shape that can be displayed in any two dimensions using the knowledge based management process subprocess, and thus the drawing elements invented automatically in series are stored and / or displayed using the knowledge base management. Based on the information stored in the character extraction process sub-process, the drawing elements extracted during the information extraction are displayed, even if the automatic identification is unsuccessful, but during the knowledge base management the character extraction is performed by training the unknown shape and thus all drawing elements of all selected shapes on the document are completely collected and displayed even if the identical shapes differ greatly in size, location or orientation. In a sub-process of character extraction process, the shape is scanned using a virtual dividing line so that it starts at the minimum marker line at the first edge of the leftmost search rectangle and ends at the maximum marker line at the rightmost closing edge of the last search rectangle according to the information extraction process, but when each character or drawing element is found, the drawing element is horizontally divided by a virtual dividing line, and if an interpretable character or symbol is obtained after this, the information extraction and display of the new drawing element and / or symbol or other trained marking is performed.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Drawing input device

    JP1992034652A

  • Symbol recognizing system

    JP1994089316A