Gap Shifting for Tabular Text Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Optical character recognition (OCR) systems face challenges in detecting text in tabular forms, particularly when there are large gaps between left-justified and right-justified text, leading to missed recognition of right-justified text.
Innovation Solution
A gap-shifting approach is employed to move text closer to other text in the row, allowing the text detection engine to re-detect previously overlooked right-justified text, which is then provided to the recognition engine for accurate processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a text detection engine operates left to right on tabular text, then left-justified text is detected accurately, but right-justified text with large gaps is missed
Solution Approach 1:
The patent applies gap shifting as a preliminary action before text detection. By shifting text elements to close gaps between left-justified and right-justified text in tabular layouts, the text detection engine can subsequently detect all text elements including previously missed right-justified text.
Solution Approach 2:
The patent introduces a new dimension of text manipulation by shifting text horizontally to close gaps. This dimensional adjustment allows the text detection engine to capture right-justified text that would otherwise be missed due to large horizontal spacing in traditional left-to-right detection.
2Loss of information
If gap shifting is applied to close text gaps, then right-justified text detection improves, but text positioning accuracy may be compromised
Solution Approach 1:
Gap shifting is performed as a preliminary transformation before text detection. The system records the shifting amount and applies the reverse transformation after detection, ensuring that text positioning accuracy is restored while still benefiting from improved text detection during the intermediate stage.
Solution Approach 2:
The system implements feedback by recording the gap shifting amount and using it to reconstruct the original text positioning after detection. This feedback mechanism ensures that the temporary positioning change for detection purposes does not compromise the final positioning accuracy of detected text.
Data Source
AI summary
Disclosed are various embodiments for improving optical character recognition approaches through the use of gap shifting. A text detection process is performed upon an image to detect a first region of text. A second region that is in line with the first region is shifted to reduce a gap between the first region and the second region, thereby creating a modified image. The text detection process is performed upon the modified image in order to detect text within the second region.