Typesetness Score for Hand-Drawn Table Geometry
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Image processing devices face difficulties in determining the geometry of hand-drawn tables with non-straight lines, making it challenging to generate a high-level representation for inclusion in electronic documents.
Innovation Solution
A method that generates a skeleton graph for the table, identifies angles and lengths of edges, and calculates a typesetness score to compare the table to a template table, allowing for the adjustment of tolerances in generating a high-level representation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If hand-drawn lines are used to create tables, then ease of creation is improved, but manufacturing precision deteriorates because the lines are rarely straight and make it difficult to determine table geometry
Solution Approach 1:
The patent transforms the hand-drawn table image into a parameterized representation by detecting line segments, calculating their angles and lengths, and comparing these parameters against ideal template table parameters. This parameter transformation allows the system to work with quantitative measurements rather than raw pixel data, enabling precise geometric analysis of hand-drawn tables while preserving the ease of hand-drawn creation.
Solution Approach 2:
The patent replaces manual geometric measurement and analysis with an automated image processing system that uses computer vision algorithms to detect lines, calculate angles and lengths, and compute typesetness scores. This substitution of mechanical/manual processes with computational methods enables precise measurement of hand-drawn table geometry without requiring straight lines or perfect drawing skills.
2Adaptability or versatility
If hand-drawn tables with non-straight lines are processed, then adaptability is improved, but measurement precision deteriorates because it is difficult to determine table geometry and generate high-level representation
Solution Approach 1:
The patent segments the hand-drawn table image into individual line segments and identifies their geometric parameters (angles and lengths) separately. By breaking down the complex hand-drawn table into discrete measurable components, the system can accurately measure each segment's properties and aggregate them to determine overall table geometry, thereby maintaining measurement precision while adapting to non-straight hand-drawn lines.
Solution Approach 2:
The patent introduces an intermediary computational model that bridges the gap between hand-drawn table images and high-level geometric representations. This intermediary system uses template matching and typesetness scoring to translate imperfect hand-drawn lines into structured geometric data, enabling accurate measurement and representation generation without requiring the input to be perfectly straight or规范的.
3Measurement precision
If template matching is used to calculate typesetness score, then measurement precision is improved, but device complexity increases due to skeleton graph generation and angle/length identification
Solution Approach 1:
The patent performs preliminary actions by pre-defining template tables with ideal geometric parameters before processing hand-drawn tables. The template matching process compares detected line parameters against these pre-established templates, which simplifies the measurement process and improves precision by providing reference standards. This preliminary preparation reduces the computational complexity during actual table analysis.
Solution Approach 2:
The patent transitions from two-dimensional image processing to a parameter space representation by extracting angles and lengths as separate dimensional attributes. This dimensional transformation allows the system to analyze table geometry in a simplified parameter space rather than working directly with complex image data, improving measurement precision while managing system complexity through dimensional decomposition.
Data Source
AI summary
A method for image processing is provided. The method includes: obtaining an image including a table; generating, for the table, a skeleton graph including a plurality of edges; identifying a plurality of angles and a plurality of lengths for the plurality of edges; and calculating a typesetness score that compares the table to a template table based on the plurality of angles and the plurality of lengths.