OCR Layout Recognition for Accurate Text Line Alignment

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional optical character recognition (OCR) technologies struggle with accurately aligning recognized characters to the layout of text in images, leading to errors in text interpretation.

Innovation Solution

A character recognition system utilizing a deep learning-based model to extract character area, inter-character space, interline scale, and orientation information, followed by a text line recognition unit to align characters based on these features, and a layout analysis unit to generate paragraph and line number information.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If conventional OCR technology is used to recognize characters in an image, then character recognition speed is improved, but layout alignment accuracy deteriorates

Engineering Contradiction:
Improvecharacter recognition speedVSAvoidlayout alignment accuracy
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The system segments the character recognition process into two independent modules: a character recognition module that extracts individual character information, and a layout analysis module that analyzes text line positions and orientations. This segmentation allows each module to specialize, maintaining high recognition speed while improving layout alignment accuracy through dedicated layout processing.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces layout information (text line positions, orientations, and character positions) as an intermediary between character recognition and text interpretation. This intermediary layer provides the spatial context needed to align recognized characters with their actual layout positions, resolving the accuracy issue without sacrificing recognition speed.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If deep learning models extract detailed character and layout information, then layout alignment accuracy is improved, but system complexity increases

Engineering Contradiction:
Improvelayout alignment accuracyVSAvoidsystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system divides the deep learning processing into separate specialized models: a character recognition model for identifying characters, and a layout analysis model for determining text line and character positions. This segmentation reduces the complexity of individual models while achieving high layout alignment accuracy through coordinated processing.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent extracts layout information (positions, orientations, text line structures) as a separate output from the character recognition process. By taking out layout analysis as an independent function, the system avoids the complexity of integrating multiple functions into a single complex model, while still achieving high layout alignment accuracy.

Inventive Principle:
Principle #2Taking out (Extraction)

3Measurement precision

If text lines are recognized based on multiple character features, then text line recognition accuracy is improved, but processing time increases

Engineering Contradiction:
Improvetext line recognition accuracyVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The system performs preliminary extraction of key layout features (character positions, text line positions, orientations) before final text line recognition. By preparing these features in advance, the recognition process can quickly match characters to text lines using pre-computed spatial information, improving accuracy without significant time penalty.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent replaces complex mechanical processing of character relationships with a data-driven approach using deep learning models that automatically learn spatial patterns. The layout analysis model processes multiple character features simultaneously through neural network operations, achieving high text line recognition accuracy more efficiently than traditional sequential processing methods.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentEP4675581A1Character recognition and document interpretation method and system based on layout recognition
Publication Date: 2026.01.07 AJOU UNIV IND ACADEMIC COOP FOUND
  • EP4675581A1 patent drawingFigure 1
  • EP4675581A1 patent drawingFigure 2
  • EP4675581A1 patent drawingFigure 3

AI summary

A character recognition system includes: a character-related information extraction unit configured to include a deep learning model trained to extract character area information, inter-character space area information, interline scale information of each character, and orientation information of each character from an image including text; a word unit division recognition unit configured to obtain word division information obtained by dividing characters included in the image into word units based on the character area information and inter-character space area information; a text line recognition unit configured to recognize text lines in the image based on the character area information, interline scale information, and orientation information; a layout analysis unit configured to obtain layout information of the text included in the image based on the recognized text lines; and a character recognition unit configured to recognize each of the character included in the image and obtain text data in which the recognized characters are aligned based on the word division information and the layout information.