Regular expressions for decoding of neural network outputs

详细信息    查看全文

• 作者：Tobias Strauß ; ; ^{tobias.strauss@uni-rostock.de" class="auth_mail" title="E-mail the corresponding author} ; Gundram Leifert ^{gundram.leifert@uni-rostock.de" class="auth_mail" title="E-mail the corresponding author} ; Tobias Grü ; ning ^{tobias.gruening@uni-rostock.de" class="auth_mail" title="E-mail the corresponding author} ; Roger Labahn ^{roger.labahn@uni-rostock.de" class="auth_mail" title="E-mail the corresponding author}
• 关键词：Regular expressions ; Decoding ; Connectionist Temporal Classification ; Handwriting recognition
• 刊名：Neural Networks
• 出版年：2016
• 出版时间：July 2016
• 年：2016
• 卷：79
• 期：Complete
• 页码：1-11
• 全文大小：617 K

文摘

This article proposes a convenient tool for decoding the output of neural networks trained by Connectionist Temporal Classification (CTC) for handwritten text recognition. We use regular expressions to describe the complex structures expected in the writing. The corresponding finite automata are employed to build a decoder. We analyze theoretically which calculations are relevant and which can be avoided. A great speed-up results from an approximation. We conclude that the approximation most likely fails if the regular expression does not match the ground truth which is not harmful for many applications since the low probability will be even underestimated. The proposed decoder is very efficient compared to other decoding methods. The variety of applications reaches from information retrieval to full text recognition. We refer to applications where we integrated the proposed decoder successfully.