Scholarlink

Document digitization is a need of the hour nowadays to preserve the precious treasure of information printed in various sources like books, journals, historical periodicals, geographical maps, etc. The digitization of any type of maps facilitates the analysis of ancient, distinctive, and important information sources. Many of the documents especially map segmentation pose many challenges in developing OCR. Many of the authors have tried to convert historical or geographical documents through OCR and successfully build archives of newspapers, books, journals, etc. In this paper, a technique has been presented based on run length smearing algorithm and projection profile to segment maps embedded in scanned printed text images which have produced high accuracy results.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A Novel Technique to Segment Maps from Scanned Document Images

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Rupinderpal Kaur,
  • Munish Kumar,
  • Manish Kumar Jindal

摘要

Document digitization is a need of the hour nowadays to preserve the precious treasure of information printed in various sources like books, journals, historical periodicals, geographical maps, etc. The digitization of any type of maps facilitates the analysis of ancient, distinctive, and important information sources. Many of the documents especially map segmentation pose many challenges in developing OCR. Many of the authors have tried to convert historical or geographical documents through OCR and successfully build archives of newspapers, books, journals, etc. In this paper, a technique has been presented based on run length smearing algorithm and projection profile to segment maps embedded in scanned printed text images which have produced high accuracy results.