Hand-Drawn Table Geometry Determination via Skeleton Graph
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Image processing devices face difficulties in accurately determining the geometry of hand-drawn tables with non-straight lines, making it challenging to generate a high-level representation for electronic documents.
Innovation Solution
A method involving identifying connected components, generating a skeleton graph, calculating a typesetness score, determining a rotation angle, and creating a custom coordinate system to accurately identify and link cells in the table, ultimately exporting a grid representation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If hand-drawn lines are used to create tables, then the table can be created freely by users, but the geometry of the table becomes difficult to determine accurately
Solution Approach 1:
The patent segments the hand-drawn table image into multiple connected components representing different table elements (lines, cells, intersections). By dividing the complex hand-drawn structure into discrete segments, the system can analyze and process each component separately, making geometry determination feasible despite the freehand drawing style.
Solution Approach 2:
The patent introduces a skeleton graph as an intermediary representation between the raw hand-drawn image and the final table geometry. The skeleton graph extracts structural relationships and connectivity patterns from the hand-drawn lines, serving as a bridge that enables accurate geometry determination while preserving user drawing freedom.
2Ease of operation
If hand-drawn lines are used for tables, then users can draw tables naturally, but generating a high-level representation becomes difficult
Solution Approach 1:
The patent performs preliminary processing of the hand-drawn table image by identifying connected components and constructing a skeleton graph before attempting to generate the high-level representation. This preliminary structuring simplifies the subsequent representation generation by organizing the data in a computationally tractable format.
Solution Approach 2:
The patent replaces manual analysis of hand-drawn tables with an automated image processing system that uses connected component analysis and skeleton graph construction. This substitution of mechanical/image processing methods for manual interpretation enables complex representation generation from natural hand-drawn input.
3Ease of manufacture
If standard image processing methods are used on hand-drawn tables, then processing is simple, but accuracy in determining table structure is poor
Solution Approach 1:
The patent employs dynamic programming or iterative optimization methods to refine the table structure determination. The system dynamically adjusts the interpretation of hand-drawn lines by considering connectivity patterns, line intersections, and spatial relationships, improving reliability while maintaining computational feasibility.
Solution Approach 2:
The patent incorporates feedback mechanisms where the skeleton graph construction and connected component analysis inform each other iteratively. The system uses feedback from the detected structure to refine the geometry determination, improving accuracy while building upon the initial simple image processing steps.
Data Source
AI summary
A method for processing an image including a table is provided. The method includes: identifying a plurality of connected components that form the table; generating a skeleton graph comprising a plurality of edges based on the plurality of connected components; calculating a typesetness score that compares the table to a template table based on the plurality of edges; determining a rotation angle for the table and generating a custom coordinate system for the table based on the rotation angle; identifying a plurality of cells in the table based on the skeleton graph, the typesetness score, and the custom coordinate system; linking each of the plurality of cells based on the plurality of edges and validating a link between the plurality of cells; generating a grid of cells based on the link and comprising the plurality of cells; and exporting the grid as a high-level representation of the table.