首页> 外国专利> SYSTEM AND METHOD FOR DATA PROCESSING, WHICH CORRECTS ERROR IN CHARACTER RECOGNITION OF DIGITAL IMAGE OF DOCUMENT FORMAT

SYSTEM AND METHOD FOR DATA PROCESSING, WHICH CORRECTS ERROR IN CHARACTER RECOGNITION OF DIGITAL IMAGE OF DOCUMENT FORMAT

机译:用于在文档格式的数字图像的字符识别中纠正错误的数据处理系统和方法

摘要

PURPOSE: To improve a character recognizing method for information in a document format image. CONSTITUTION: Recognition coded data are inputted to a coded data buffer part of a field data segment and recognition error data are inputted to an error buffer part. A following correcting process is applied to the recognition code data, and performed by reinforcing correction segments corresponding to respective character strings to be corrected into MGDS(machine generated data structure). The sequence of the correction stage is applied to a specific character string and in each correction step, another correction segment is added to the MGDS. In respective stages of the correction, the optimum prediction of the character strings is arranged at the coded data buffer part of the field data segment. Consequently, the optimum prediction of the information contents of a document field becomes immediately usable in the respective stages of the correction of the data processing system and for final use.
机译:目的:改进文档格式图像中信息的字符识别方法。组成:识别编码数据输入到字段数据段的编码数据缓冲区,识别错误数据输入到错误缓冲区。随后的校正处理应用于识别码数据,并且通过将与要校正的各个字符串相对应的校正段增强为MGDS(机器生成的数据结构)来执行。校正阶段的序列被应用于特定的字符串,并且在每个校正步骤中,另一个校正段被添加到MGDS。在校正的各个阶段中,在字段数据段的编码数据缓冲器部分处布置字符串的最佳预测。因此,在数据处理系统的校正的各个阶段以及最终使用中,立即可以使用对文档字段的信息内容的最佳预测。

著录项

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号