주황색 웨이브 모양의 배경

Textscope®

Doc Parser

주황색 웨이브 모양의 배경

Textscope®

Doc Parser

Company

Textscope®

Business

Insight

Textscope® Doc Parser: Document Layout Analysis Solution​

Textscope® Doc Parser is a powerful solution for analyzing document layouts, detecting paragraphs, images, tables, and other elements across diverse formats. It transforms both structural and non-textual information into actionable data, optimizing document usability and maximizing data value.​

Key features of Doc Parser

Analysis and extraction of document and table structures

Analysis and extraction of document and table structures

Analysis and extraction of document and table structures

Recognition of over 10 different layout elements

Recognition of over 10 different layout elements

Document titles, section subtitles

Document titles, section subtitles

Text paragraphs, lists, equations

Text paragraphs, lists, equations

Tables, pictures, captions

Tables, pictures, captions

Header, footer, footnote

Header, footer, footnote

Arrangement of layout elements in a natural reading order​

Arrangement of layout elements in a natural reading order​

Detection of font sizes, image dimensions, and location​

Detection of font sizes, image dimensions, and location​

Relationships between figures/tables and their captions​

Relationships between figures/tables and their captions​

Table structure recognition

Table structure recognition

Table structure recognition

Table detection across various styles​

Table detection across various styles​

Identifies table layout elements using HTML tags like <table>, <thead>, <td>, and more

Identifies table layout elements using HTML tags like <table>, <thead>, <td>, and more

Table caption detection​

Table caption detection​

Supports recognition of merged cells​

Supports recognition of merged cells​

Header Information Recognition

Header Information Recognition

Other features of Doc Parser

Image file recognition

Image file recognition

Recognize document elements such as layout, text, table, etc. in image files such as scan/fax documents

Recognize document elements such as layout, text, table, etc. in image files such as scan/fax documents

Works on low-quality images such as shadows, noise, and photographic angles

Works on low-quality images such as shadows, noise, and photographic angles

Exceptional accuracy in both printed and handwritten text recognition​

Exceptional accuracy in both printed and handwritten text recognition​

AI-powered layout analysis for accurate image recognition​

AI-powered layout analysis for accurate image recognition​

Supports a variety of input/output file formats

Supports a variety of input/output file formats

Input

Input

Office documents: PDF, Hangul (hwp, hwpx), Word (doc, docx), PowerPoint (ppt, pptx), Excel (xls, xlsx)

Office documents: PDF, Hangul (hwp, hwpx), Word (doc, docx), PowerPoint (ppt, pptx), Excel (xls, xlsx)

Image documents: JPG, PNG, TIFF, BMP, GIF, PDF, etc

Image documents: JPG, PNG, TIFF, BMP, GIF, PDF, etc

Output

Output

Text formats: export as html, markdown, text, and other text-based formats

Text formats: export as html, markdown, text, and other text-based formats

Table exports: extract tables and export them as excel or csv files

Table exports: extract tables and export them as excel or csv files

Image exports: save extracted images as standalone files (jpg, png, etc.)

Image exports: save extracted images as standalone files (jpg, png, etc.)

LLM/RAG services and integration ready

LLM/RAG services and integration ready

Enhanced Information Delivery: Elevates RAG and LLM search and response accuracy with richer document data.

Enhanced Information Delivery: Elevates RAG and LLM search and response accuracy with richer document data.

Data Integration for Vector Embedding: Streamlines data connections for vector embedding

Data Integration for Vector Embedding: Streamlines data connections for vector embedding

Customized Data Formatting: Tailors data formats to boost LLM performance.

Customized Data Formatting: Tailors data formats to boost LLM performance.

Doc Parser can be utilized as follows.

Doc Parser example

01

01

Choose a document

Choose a document

It supports office documents (PDF, Hangul, Word, PowerPoint, Excel, etc.) and image documents (JPG, PNG, TIFF, BMP, GIF, PDF, etc.).

It supports office documents (PDF, Hangul, Word, PowerPoint, Excel, etc.) and image documents (JPG, PNG, TIFF, BMP, GIF, PDF, etc.).

02

02

Document Parsing

Document Parsing

Recognizes objects in a document, such as text, pictures, and tables.

Recognizes objects in a document, such as text, pictures, and tables.

03

03

The result

The result

The recognition result is converted into structured data such as HTML, Markdown, and Text, and the extracted image can be stored as a separate file (jpg, png, etc.).

The recognition result is converted into structured data such as HTML, Markdown, and Text, and the extracted image can be stored as a separate file (jpg, png, etc.).

04

04

Application

Application

Interworks with RAG/LLM services through Vector Embedding. You can also parse the contents of a document and automatically convert it into a web page for mobile or PC (html) to provide the information contained in the document over the web.

Interworks with RAG/LLM services through Vector Embedding. You can also parse the contents of a document and automatically convert it into a web page for mobile or PC (html) to provide the information contained in the document over the web.

문서를 보는 남자

Interested in learning more about

Doc Parser?

Please contact us right away. A Document AI specialist will provide you
with the best ways to enhance the value of your document data
as quickly and comprehensively as possible.

문서를 보는 남자

Interested in learning more

about

Doc Parser?

Please contact us right away. A Document AI specialist will provide you with the best ways to enhance the value of your document data as quickly and comprehensively as possible.

7th Floor, JBI Building, 10 Bangbaechun-ro 2-gil, Seocho-gu, Seoul

Product & Technical Consulting

T. +82) 2 6289 0501

General Inquiry

T. +82) 2 6331 1853

Copyright © 2024 Lomin.ai. All Rights Reserved

7th Floor, JBI Building, 10 Bangbaechun-ro 2-gil, Seocho-gu, Seoul

Product & Technical Consulting

General Inquiry

Copyright © 2024 Lomin.ai. All Rights Reserved