公开(公告)号：US20050259866A1

公开(公告)日：2005-11-24

申请号：US10850335

申请日：2004-05-20

Applicant: Charles Jacobs ,  James Rinker ,  Patrice Simard ,  Paul Viola

Inventor： Charles Jacobs ,  James Rinker ,  Patrice Simard ,  Paul Viola

IPC: G06K9/20 ,  G06K9/34 ,  G06K9/46 ,  G06K9/66 ,  G06K9/72 ,  G06N3/00 ,  G06K9/62 ,  G06K9/18

CPC classification number: G06K9/344 ,  G06K9/00463 ,  G06K9/4614 ,  G06K9/4628 ,  G06K9/723 ,  G06K2209/01

Abstract: A global optimization framework for optical character recognition (OCR) of low-resolution photographed documents that combines a binarization-type process, segmentation, and recognition into a single process. The framework includes a machine learning approach trained on a large amount of data. A convolutional neural network can be employed to compute a classification function at multiple positions and take grey-level input which eliminates binarization. The framework utilizes preprocessing, layout analysis, character recognition, and word recognition to output high recognition rates. The framework also employs dynamic programming and language models to arrive at the desired output.

Abstract translation: 低分辨率拍摄文档的光学字符识别（OCR）的全局优化框架，将二值化类型过程，分割和识别结合到一个过程中。 该框架包括对大量数据进行培训的机器学习方法。 可以采用卷积神经网络来计算多个位置的分类函数，并采用消除二值化的灰度级输入。 该框架利用预处理，布局分析，字符识别和字识别来输出高识别率。 该框架还采用动态编程和语言模型来达到所需的输出。