Inverse Video extraction in Scribo

From LRDE

Revision as of 17:05, 9 January 2018 by Bot (talk | contribs) (Created page with "{{CSIReport | authors = Coddy Levi | title = Inverse Video extraction in Scribo | year = 2011 | abstract = Text in document images that is subject to extraction through docume...")
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)

Abstract

Text in document images that is subject to extraction through document analysis can be present in two forms: dark text over light background (conventional form) or light text over dark background (Inverse Video). This report introduces the problematic of extracting inverse video text using an existing toolchain, its integration in Scribo, the problems it involves and solutions to better results that we have explored.