Handwritten Document Layout Preservation via Line Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current OCR techniques fail to accurately recognize and preserve the layout and composition of handwritten documents, leading to unnatural line breaks and loss of document structure when converting handwritten characters into character codes.
Innovation Solution
An electronic apparatus with a line recognition module, character recognition module, and generator that recognizes lines and character codes in handwritten documents, and generates formed document data by maintaining the original layout and composition, including features like line breaks and font sizes, to produce a readable and structured digital output.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If OCR technique is used to convert handwritten characters into character codes, then character recognition is achieved, but layout and composition information is lost
Solution Approach 1:
The patent segments the handwritten document into multiple lines before character recognition. The line recognition unit divides the document image into line regions based on spatial distribution of handwritten content, preserving the original line breaks and layout structure. This segmentation approach maintains the compositional information while enabling accurate character recognition within each line context.
2Productivity
If characters are recognized in turn from upper left position, then character codes are output in recognition order, but original document structure is not preserved
Solution Approach 1:
The patent performs preliminary line recognition and segmentation before character recognition. By pre-identifying line regions and their spatial positions, the system establishes the document structure in advance. This preliminary action allows subsequent character recognition to proceed efficiently within each line while automatically preserving the original layout, eliminating the need for post-processing reorganization.
3Productivity
If handwritten document is converted to character codes without line recognition, then conversion speed is fast, but line breaks and paragraph structure are lost
Solution Approach 1:
The patent implements segmentation at the line level by dividing the handwritten document into distinct line regions based on spatial analysis. This segmentation preserves line breaks and paragraph structures while enabling parallel processing of character recognition across multiple lines, maintaining both structural integrity and conversion efficiency.
4Loss of information
If layout analysis is performed to preserve document structure, then composition information is maintained, but processing complexity increases
Solution Approach 1:
The patent uses simple spatial segmentation based on the distribution of handwritten content to identify line regions. This approach preserves composition information through straightforward geometric analysis rather than complex layout algorithms, maintaining low processing complexity while achieving effective structure preservation.
Solution Approach 2:
The patent applies local quality analysis by examining the spatial distribution characteristics within different regions of the handwritten document. By analyzing local patterns of handwriting density and position, the system identifies line boundaries and structural elements without requiring global complex processing, thus preserving composition with minimal added complexity.
Data Source
AI summary
According to one embodiment, an electronic apparatus includes a line recognition module, a character recognition module and a generator. The line recognition module recognizes lines in a handwritten document. The character recognition module recognizes character codes corresponding to handwritten characters in a first line and a second line which follows the first line. The generator generates, if the first and second lines satisfy a condition, document data using first character codes corresponding to the first line and second character codes corresponding to the second line, the formed document data including either one of the first character codes at a position of the second line or including at least one of the second character codes at a position of the first line.