From owner-chemistry@ccl.net Fri Apr  9 10:10:01 2010
From: "Igor Filippov Contr igorf-#-helix.nih.gov" <owner-chemistry\a/server.ccl.net>
To: CCL
Subject: CCL: large validation set for chemical OCR software
Message-Id: <-41632-100408175448-29246-9rDjRORZGea7S+n2+5oVKg\a/server.ccl.net>
X-Original-From: "Igor Filippov [Contr]" <igorf[A]helix.nih.gov>
Content-Transfer-Encoding: 7bit
Content-Type: text/plain; charset="UTF-8"
Date: Thu, 08 Apr 2010 17:51:15 -0400
Mime-Version: 1.0


Sent to CCL by: "Igor Filippov [Contr]" [igorf[*]helix.nih.gov]

Dear Colleagues,

We would like to announce immediate availability of a large set of
molecular structure images plus associated MOL files, put together in
the context of development, testing and validation of chemical optical
structure recognition software. It is based on US Patent Office Complex
Work Units and can be used, e.g., for testing of software such as OSRA,
ChemReader, CLiDE, and ChemOCR. It comprises 5735 images and
ground-truth MOL files, which makes it the largest such validation set
available to the public to the best of our knowledge.  It is available
for download at:
http://cactus.nci.nih.gov/osra
http://osra.sourceforge.net/

We would like to thank Dr. Steve Boyer and Dr. John Kinney for
collaboration in developing the validation set.

Sincerely,
Igor Filippov and Marc Nicklaus