Table Coordinate Correction for Accurate Document Table Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional document recognition technologies fail to accurately calculate table coordinate information, leading to errors in table form recognition and incorrect electronic document production, particularly in Excel documents, due to issues like data loss and image distortion.
Innovation Solution
A method and apparatus for correcting table coordinate information by producing lists of first and second coordinate elements, grouping them based on size and position, and adjusting the coordinate information using representative values to minimize computational resources and improve accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If conventional machine learning techniques are used to extract table cells from document images, then cell extraction can be performed automatically, but coordinate information errors occur due to probability-based processing
Solution Approach 1:
The patent applies feedback by using the extracted coordinate information to identify errors and correct them. The system processes coordinate data, detects inconsistencies through statistical analysis, and iteratively refines the coordinate information to improve accuracy while maintaining automatic extraction.
Solution Approach 2:
The patent changes parameters by transforming raw coordinate data into corrected coordinate information through statistical processing. It modifies the coordinate values based on identified error patterns and uses parameter adjustments to resolve inconsistencies in the extracted table structure.
2Measurement precision
If coordinate information is processed to correct errors, then table structure accuracy improves, but computational resources increase
Solution Approach 1:
The patent applies partial action by selectively correcting only the coordinate information that contains errors rather than processing all coordinate data uniformly. It identifies specific problematic areas in the table structure and applies correction only where needed, reducing overall computational burden.
Solution Approach 2:
The patent uses parameter changes to optimize the correction process by adjusting processing thresholds and selection criteria. It modifies parameters such as error detection sensitivity and correction application rules to balance accuracy improvement with computational efficiency.
3Measurement precision
If all coordinate elements are processed individually, then correction precision may improve, but processing time increases
Solution Approach 1:
The patent merges coordinate elements by grouping them based on their spatial relationships and functional roles in the table structure. It combines related coordinate data points into sets that can be processed together, maintaining correction precision while reducing the total number of individual processing operations.
Solution Approach 2:
The patent segments the coordinate processing into distinct phases: extraction, error identification, and correction. It divides the coordinate data into logical groups corresponding to different table structure elements, allowing efficient processing of each segment with appropriate methods.
Data Source
AI summary
The present disclosure relates to a method, an apparatus, a system, and a computer program for correcting table coordinate information. The present disclosure discloses a method for correcting table coordinate information performed by one or more processors in an apparatus, which may include: producing a list of first coordinate elements for a first axis or second coordinate elements for a second axis on the basis of coordinate information of a plurality of cells included in a table of a document image; performing grouping on the list of first coordinate elements or the list of second coordinate elements; and correcting the coordinate information of the cells of the table on the basis of the first coordinate elements or the second coordinate elements on which the grouping was performed.