Document Image Character Recognition Enlargement Unit
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing character recognition methods, such as those described in Japanese Patent Laid-Open No. 2019-159633, face challenges in enhancing accuracy for characters of different sizes, as size compression can damage information and lead to inaccurate recognition.
Innovation Solution
An image processing apparatus that obtains document images, identifies and enlarges target text line regions with unique height-width ratios to match those of ordinary characters, allowing for accurate character recognition without damaging the character information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If the size of second characters is compressed to match the reference size, then the character size uniformity is improved, but the character information may be damaged leading to reduced recognition accuracy
Solution Approach 1:
Instead of compressing large characters to match the reference size, the patent inverts the approach by enlarging the reference characters to match the size of the second characters. This inversion allows the larger characters to maintain their original information while the reference characters are scaled up, thereby avoiding information loss and improving recognition accuracy.
Solution Approach 2:
The patent changes the size parameter of the reference characters dynamically based on the detected character sizes in the document. By adjusting the reference character size to match the second characters, the system adapts to different document formats and character size variations without compromising the integrity of the original character information.
2Stability of the object's composition
If compression processing is applied to second characters, then the character size consistency is improved, but the information on the character may be damaged
Solution Approach 1:
The patent inverts the conventional compression approach by implementing enlargement processing instead. Reference characters are enlarged to match the size of second characters, thereby maintaining the original information content of the second characters while achieving size consistency through scaling up rather than scaling down.
Solution Approach 2:
The patent performs preliminary detection of character sizes in the document before processing. By identifying the presence of second characters and their size characteristics in advance, the system can pre-determine the appropriate enlargement ratio for reference characters, ensuring that the enlargement process maintains information integrity while achieving the desired size consistency.
Data Source
AI summary
An apparatus includes: an obtainment unit configured to obtain a document image at least including first character images and second character images, the first character images being ordinary character images appropriate for character recognition processing, the second character images having a different height-width ratio from a height-width ratio of the ordinary character images and having a greater size than a size of the ordinary character images; an enlargement unit configured to enlarge a target text line region including the second character images such that the height-width ratio of each of the second character images included in the target text line region is equal to the height-width ratio of the first character images; and a character recognition unit configured to perform character recognition processing on the target text line region enlarged by the enlargement unit.