Splitting Merged Table Cells via Skeleton Graph Confidence
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Hand-drawn tables with irregular lines pose challenges for image processing devices to determine table geometry and generate a high-level representation, as existing methods struggle to accurately split merged cells and align with electronic document formats.
Innovation Solution
A method and system that utilize a skeleton graph to identify non-corner vertices, calculate confidence values for splitlines, and generate artificial edges to split merged cells, aligning with the custom coordinate system and enabling accurate representation in electronic documents.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If hand-drawn lines are used to create tables, then the tables can be freely drawn and annotated, but the lines become irregular and curved making it difficult to determine table geometry and cell positions
Solution Approach 1:
The patent segments the hand-drawn table image into discrete geometric elements by detecting lines, curves, and intersections, then represents them as a structured graph model with vertices and edges. This segmentation transforms the irregular hand-drawn lines into measurable geometric components that can be precisely analyzed and converted into electronic document formats.
Solution Approach 2:
The patent introduces an intermediate geometric model (graph representation with vertices, edges, and intersection points) that serves as a mediator between the hand-drawn table image and the final electronic document representation. This intermediate model captures the geometric essence of hand-drawn tables while providing a structured format suitable for precise measurement and conversion.
2Productivity
If existing image processing methods are used on hand-drawn tables, then processing can be performed, but the methods struggle to accurately split merged cells and align with electronic document formats
Solution Approach 1:
The patent performs preliminary actions by first detecting all lines and curves in the hand-drawn table, then identifying intersection points and constructing a complete graph representation before attempting to split cells or generate electronic document formats. This preliminary geometric analysis ensures that all geometric relationships are captured before the splitting operation, improving accuracy.
Solution Approach 2:
The patent changes parameters by transforming the representation from pixel-based image data to geometric parameters (line equations, intersection coordinates, angles), and then to structural parameters (graph nodes, edges, cell definitions). This parameter transformation enables precise control over cell splitting operations and accurate alignment with electronic document formats.
Data Source
AI summary
A method for image processing is provided. The method includes: obtaining an image including a table that includes a plurality of cells that each includes a first border, a second border opposite the first border, a third border, and a fourth border opposite the third border; identifying, on a skeleton graph generated for the table, a first non-corner vertex on the first border of a first cell of the plurality of cells; determining a first plurality of splitlines extending from the first non-corner vertex and intersecting with the second border of the first cell without intersecting with the third or fourth border of the first cell; calculating a first plurality of confidence values for the first plurality of splitlines; and splitting the first cell by generating a first artificial edge, based on the first plurality of confidence values and a splitline of the first plurality of splitlines.