Scholarlink

The Indian Constitution recognizes Telugu, Tamil, Malayalam, and Kannada as significant languages. Approximately 90 million people worldwide speak Telugu, a language from South India. Telugu OCR, or optical character recognition, has several uses, such as enabling digital books and unstructured documents, which in turn improves communication between people. In contrast to regional languages like Telugu, Tamil, Malayalam, etc., OCR systems are extremely well trained for international languages like English and German. Telugu has a large number of different characters, which presents a significant obstacle to the development of OCR. In this study, we address this in two ways: (i) a database containing handwritten Telugu characters and (ii) an enhanced CNN model designed to identify scanned Telugu characters.

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

In-Depth Analysis and Challenges of Handwritten Telugu Character Recognition with an Improved CNN Model for Solutions

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Dumpal Koteswararao,
  • Nagaratna P. Hedge

摘要

The Indian Constitution recognizes Telugu, Tamil, Malayalam, and Kannada as significant languages. Approximately 90 million people worldwide speak Telugu, a language from South India. Telugu OCR, or optical character recognition, has several uses, such as enabling digital books and unstructured documents, which in turn improves communication between people. In contrast to regional languages like Telugu, Tamil, Malayalam, etc., OCR systems are extremely well trained for international languages like English and German. Telugu has a large number of different characters, which presents a significant obstacle to the development of OCR. In this study, we address this in two ways: (i) a database containing handwritten Telugu characters and (ii) an enhanced CNN model designed to identify scanned Telugu characters.