-
1.发明申请公开(公告)号：US20210034849A1
公开(公告)日：2021-02-04
申请号：US17075511
申请日：2020-10-20
申请人： Kodak Alaris Inc.
摘要： Embodiments of the present disclosure include a method that obtains a digital image. The method includes extracting a word block from the digital image. The method includes processing the word block by evaluating a value of the word block against a dictionary. The method includes outputting a prediction equal to a common word in the dictionary when a confidence factor is greater than a predetermined threshold. The method includes processing the word block and assigning a descriptor to the word block corresponding to a property of the word block. The method includes processing the word block using the descriptor to prioritize evaluation of the word block. The method includes concatenating a first output and a second output. The method includes predicting a value of the word block.
-
2.发明申请公开(公告)号：US20210034850A1
公开(公告)日：2021-02-04
申请号：US17075513
申请日：2020-10-20
申请人： Kodak Alaris Inc.
摘要： Embodiments of the present disclosure include a method that obtains a digital image. The method includes extracting a word block from the digital image. The method includes processing the word block by evaluating a value of the word block against a dictionary. The method includes outputting a prediction equal to a common word in the dictionary when a confidence factor is greater than a predetermined threshold. The method includes processing the word block and assigning a descriptor to the word block corresponding to a property of the word block. The method includes processing the word block using the descriptor to prioritize evaluation of the word block. The method includes concatenating a first output and a second output. The method includes predicting a value of the word block.
-
公开(公告)号：US20210103777A1
公开(公告)日：2021-04-08
申请号：US17126745
申请日：2020-12-18
申请人： Kodak Alaris Inc.
发明人： Chi ZHANG , Raymond PTUCHA , Alexander LOUI , Carl SALVAGGIO
摘要： Embodiments of the present disclosure relate to a recurrent framework based on Recurrent Highway Networks (RHNs) for sequence modeling using batch normalization. In certain embodiments, constraints within the RHNs are relaxed to reduce or avoid gradient vanishing or exploding by normalizing the current transition units in highway layers.
-
公开(公告)号：US20180322368A1
公开(公告)日：2018-11-08
申请号：US15968012
申请日：2018-05-01
申请人： Kodak Alaris Inc.
发明人： Chi ZHANG , Raymond PTUCHA , Alexander LOUI , Carl SALVAGGIO
CPC分类号： G06K9/6262 , G06K9/6256 , G06K9/6274 , G06N3/0445 , G06N3/0454 , G06N3/08 , G06N3/084
摘要： Embodiments of the present disclosure relate to a recurrent framework based on Recurrent Highway Networks (RHNs) for sequence modeling using batch normalization. In certain embodiments, constraints within the RHNs are relaxed to reduce or avoid gradient vanishing or exploding by normalizing the current transition units in highway layers.
-
5.发明申请公开(公告)号：US20180137350A1
公开(公告)日：2018-05-17
申请号：US15812681
申请日：2017-11-14
申请人： KODAK ALARIS INC.
CPC分类号： G06K9/00422 , G06F17/2735 , G06K9/00416 , G06K9/00463 , G06K9/3283 , G06K9/4628 , G06K9/627 , G06K2209/01 , G06N3/0445 , G06N3/0454 , G06N3/084
摘要： Embodiments of the present disclosure include a method that obtains a digital image. The method includes extracting a word block from the digital image. The method includes processing the word block by evaluating a value of the word block against a dictionary. The method includes outputting a prediction equal to a common word in the dictionary when a confidence factor is greater than a predetermined threshold. The method includes processing the word block and assigning a descriptor to the word block corresponding to a property of the word block. The method includes processing the word block using the descriptor to prioritize evaluation of the word block. The method includes concatenating a first output and a second output. The method includes predicting a value of the word block.
-
-
-
-
搜索结果
5 条记录 (耗时 0.015 秒)
