Skeleton Graph Repair for Hand-Drawn Table Geometry
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Hand-drawn tables in images often have imperfections like dropouts and irregular lines, making it difficult for image processing devices to determine the geometry and generate a high-level representation for electronic documents.
Innovation Solution
A method and system that generate a skeleton graph for the table, identify terminal vertices, and repair the table by adding artificial edges between terminal vertices and targets within a region of interest, aligning with the table's geometry to create a more accurate representation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If hand-drawn tables are processed directly, then the original information is preserved, but the geometry determination and high-level representation generation become difficult
Solution Approach 1:
The patent segments the hand-drawn table into discrete skeleton graph components (vertices and edges). Each terminal vertex is identified and processed independently, with regions of interest defined around each vertex to locate connection targets. This segmentation transforms the complex continuous hand-drawn lines into discrete manageable elements that can be systematically repaired.
Solution Approach 2:
The patent introduces an intermediary skeleton graph representation between the original hand-drawn table and the final high-level representation. The skeleton graph serves as a mediator that captures the table's geometric structure in a simplified form, making it easier to determine geometry and generate electronic document representations while preserving essential information.
2Manufacturing precision
If artificial edges are added to repair terminal vertices, then the table geometry accuracy is improved, but the processing steps increase
Solution Approach 1:
The patent applies local quality by focusing repair operations only on terminal vertices and their surrounding regions of interest, rather than processing the entire table uniformly. Each terminal vertex receives targeted attention with artificial edges added only where geometric inaccuracies exist, preserving processing efficiency while improving local geometry accuracy.
Solution Approach 2:
The patent performs preliminary actions by first generating the skeleton graph and identifying all terminal vertices before executing repair operations. Regions of interest are pre-defined around each terminal vertex, and target locations are predetermined based on the skeleton graph structure. This preliminary preparation enables efficient execution of repair operations without requiring complex real-time decision-making.
Data Source
AI summary
A method for image processing is provided. The method includes: obtaining an image including a table; generating a skeleton graph associated with the table and including a plurality of edges; identifying a first terminal vertex on the plurality of edges; generating a first region of interest for the first terminal vertex; determining a first target within the first region of interest; and repairing the table by adding a first artificial edge between the first terminal vertex and the first target to the skeleton graph.