Floating Form Processing via Topological Structure Alignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Processing semi-structured (floating) forms presents challenges as conventional image processing algorithms struggle to extract user-filled data due to varying form structures, requiring fixed constraints that do not apply to forms with expandable input areas or altered table structures.
Innovation Solution
A method that extracts and matches topological structures between empty and filled forms, using OCR to segment text, generate settings files, and align horizontal and vertical sequences to identify user-filled data without relying on fixed form structures, allowing for the extraction of data from forms with flexible layouts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional image processing algorithms are used to extract user-filled data from forms, then the extraction process is simple for fixed forms, but the method fails for semi-structured forms with varying layouts
Solution Approach 1:
The patent segments the form processing task into multiple components: extracting form field definitions, identifying user-filled data regions, and matching topological structures. This segmentation allows the system to handle both fixed and semi-structured forms by processing each component independently and integrating results.
Solution Approach 2:
The patent introduces a topological structure dimension to form processing. Instead of relying solely on fixed spatial coordinates, the system creates a topological representation that captures the hierarchical and relational structure of form fields, enabling matching across different form layouts while maintaining structural semantics.
2Adaptability or versatility
If fixed constraints are applied to form processing, then processing is easier for structured forms, but the method cannot handle forms with expandable input areas or altered table structures
Solution Approach 1:
The patent changes the parameter space from fixed coordinates to topological relationships. By representing form structures in terms of hierarchical relationships and spatial connections rather than absolute positions, the system can accurately extract data from forms with varying layouts, expandable fields, and dynamic table structures while maintaining extraction precision.
3Ease of operation
If the empty and filled forms are aligned based on fixed positions, then alignment is straightforward for fixed forms, but the alignment fails when form structures differ between empty and filled versions
Solution Approach 1:
The patent introduces topological structure as an intermediary between empty and filled forms. Instead of directly aligning forms based on fixed positions, the system first extracts and matches their topological structures, then uses this matched topology to guide the alignment process. This intermediary representation enables accurate alignment even when forms have different layouts or added/removed fields.
Data Source
AI summary
A method processes scanned floating forms by comparing topological structures of an empty form and a corresponding filled form. The topological structure of each form includes vertical and horizontal ordered sequences of text phrases in the form, which describe directional relationships among text phrases but not their distances. A partial structure alignment method is used in the comparison, where insertions of text in the topological structure of the filled form relative to that of the empty form is not penalized, but deletions and substitutions are penalized. Based on this comparison, some text phrases in the filled form are matched to the those in the empty form, and unmatched text in the filled form is deemed user-filled data. The method further associates user-filled data with fields of the form based on a settings file of the empty form.