📜 ⬆️ ⬇️

ABBYY Recognition Server in the service of Her Majesty's nerds

image
The Royal Botanic Garden of Edinburgh (KSE) can be called floristic MI6 - its employees grow and study plants from all over the world. Over 300 years of external “plant” intelligence, KSBE studied two thirds of the world's flora and made a truly unique botanical collection with which florist scientists and botanists and enthusiasts could work only in Edinburgh.

Recently, the botanical garden management decided to digitize and put on the Internet intelligence about 3 million plants. Botanists needed an automated solution for streaming document input and creating an electronic archive that scales easily and is highly productive. According to the recommendations of the National Library of Great Britain, the choice was made in favor of ABBYY Recognition Server .


“Do you like to collect herbarium? And process? "
')
The task was complicated by the variety of fonts, the presence of bar codes and handwriting on outdated versions of several languages ​​- some records are dated to 1690 and difficult to recognize. Cards with intelligence about each plant are full of numbers, first and last names and abbreviations - like this:

image

In addition, KBSE has its own image management system, where TIFF files of all cards and notes are stored. So, besides the quality of the digitized material and the automation of the text processing process, the garden's management wanted Recognition Server to harmonize harmoniously with this system.

Herbarium Online Project

And they worked together.
The following happens in the black-black box:
Recognition Server has access to all TIFFs that are stored in one of the folders of the image management system. After processing, the program creates two files - a PDF with the ability to search (just in case of a fire) and a simple text file. The latter is sent to a special folder located on the TSF server, and there their own program catches it - it adds this file to the MySQL database.

Thus, the digitized herbarium is now available through the KBSE website in the Herbarium Catalog section with the ability to search by several parameters.

It looks like this:

We are looking, for example, the most ordinary buttercup. The Lingvo dictionary tells you what to look for. Ranunculus gen.

image

The service will issue a large list of cards for all samples of buttercups imported to the UK from various countries by various expeditions. Pictures of dried samples are attached. This is a small part of the list. Choose a card plants from Israel with the image.

image

The picture can be increased. For interest, see the accompanying text:

image

Now florist scientists and botanists-enthusiasts can find the material they need from anywhere in the world.

Elena Agafonova,
translator

Source: https://habr.com/ru/post/116330/


All Articles