- 【Updated on May 12, 2025】 Integration of CiNii Dissertations and CiNii Books into CiNii Research
- Trial version of CiNii Research Knowledge Graph Search feature is available on CiNii Labs
- 【Updated on June 30, 2025】Suspension and deletion of data provided by Nikkei BP
- Regarding the recording of “Research Data” and “Evidence Data”
Correction of Japanese OCR Output with a coded Processing Method.
-
- Miyahara Sueharu
- Human Interface Laboratory Nippon Telegraph and Telephone Corporation
-
- Suzuki Akira
- Human Interface Laboratory Nippon Telegraph and Telephone Corporation
Bibliographic Information
- Other Title
-
- 日本語OCRにおける認識結果の一括修正
Search this article
Description
To improve the productivity of data input by using a Japanese OCR,the extraction and correction of rejected and erroneouscharacters must be made more efficient. To input correct data efficiently,this paper proposes a one- stroke multi-level correction process.The character′s image and ca ndidates for each category.are shown at the same time to minimize the confirmation and modification overhead.Compared to the technique of character by character correction,this approach reduces the required tasks by 68% in an experiment with 32,490 printed kanji characters.
Journal
-
- IEICE technical report. Pattern recognition and understanding
-
IEICE technical report. Pattern recognition and understanding 93 (228), 41-48, 1993-09-16
The Institute of Electronics, Information and Communication Engineers
- Tweet
Keywords
Details 詳細情報について
-
- CRID
- 1573387452154506880
-
- NII Article ID
- 110003299839
-
- NII Book ID
- AN10013232
-
- Text Lang
- ja
-
- Data Source
-
- CiNii Articles