{"id":2747,"date":"2011-10-11T13:13:13","date_gmt":"2011-10-11T13:13:13","guid":{"rendered":"http:\/\/impactocr.wordpress.com\/?p=599"},"modified":"2011-10-11T13:13:13","modified_gmt":"2011-10-11T13:13:13","slug":"werkzeuge-zur-bildoptimierung","status":"publish","type":"post","link":"https:\/\/digitisation.eu\/?p=2747","title":{"rendered":"Werkzeuge zur Bildoptimierung"},"content":{"rendered":"<p>Gerd Zechmeister of the Austrian National Library talked about different approaches to improve OCR accuracy by working on the many challenges inherent in the digital images of historical texts (e.g. geometrical correction against curves and creases; removal of black borders around pages, better binarisation, &#8230;).<\/p>\n<p><!--more--><\/p>\n<p>http:\/\/vimeo.com\/32259328<\/p>\n[slideshare id=9696876&amp;doc=04zechmeisterbildoptimierung-111014091135-phpapp02]\n<p>Gerd Zechmeister von der\u00a0\u00d6sterreichischen Nationalbibliothek besch\u00e4ftigte sich mit der Bildoptimierung, also der Bearbeitung der Bilder auf eine Texterkennung hin. Denn ein optisch problemloses Bild lbedeutet noch nicht automatisch eine gute Texterkennung. Er stellte kurz einige der Herausforderungen des Materials vor (W\u00f6lbungen und Knicke im Papier, schwankendes Druckbild, durchscheinender Text von der Seitenr\u00fcckseite, schwarze R\u00e4nder), bevor er an Hand einiger Beispielbilder auf m\u00f6gliche L\u00f6sungen einging, darunter Randentfernung, geometrische Korrektur und verbesserte Binarisierung. Dabei wies er auch daraufhin, dass die ideale Reihenfolge nicht von vornherein feststeht, sondern experimentell herausgefunden werden muss ( F\u00fchrt bspw. Randentfernung vor oder nach Binarisierung zu besseren OCR-Ergebnissen?)<\/p>\n<p>F\u00fcr seinen sp\u00e4teren Vortrag &#8220;Verbesserte OCR-Software f\u00fcr historische Dokumente&#8221; (um 15:10 Uhr),\u00a0k\u00fcndigte er den Vergleich von OCR-Ergebnissen mit und ohne Bildvorverarbeitung an.<\/p>\n<p><em>RM (BSB); Mark-Oliver Fischer (BSB)<br \/>\n<\/em><\/p>\n","protected":false},"excerpt":{"rendered":"<p>Gerd Zechmeister of the Austrian National Library talked about different approaches to improve OCR accuracy by working on the many challenges inherent in the digital images of historical texts (e.g. geometrical correction against curves and creases; removal of black borders around pages, better binarisation, &#8230;).<\/p>\n","protected":false},"author":4217,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[154],"tags":[],"class_list":["post-2747","post","type-post","status-publish","format-standard","hentry","category-discussions"],"acf":[],"_links":{"self":[{"href":"https:\/\/digitisation.eu\/index.php?rest_route=\/wp\/v2\/posts\/2747","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/digitisation.eu\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/digitisation.eu\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/digitisation.eu\/index.php?rest_route=\/wp\/v2\/users\/4217"}],"replies":[{"embeddable":true,"href":"https:\/\/digitisation.eu\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=2747"}],"version-history":[{"count":0,"href":"https:\/\/digitisation.eu\/index.php?rest_route=\/wp\/v2\/posts\/2747\/revisions"}],"wp:attachment":[{"href":"https:\/\/digitisation.eu\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=2747"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/digitisation.eu\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=2747"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/digitisation.eu\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=2747"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}