Document Table Shape Detection for Rounded Corner Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional OCR methods struggle to accurately recognize and determine the shape of tables which have rounded or curved corners may not be recognized and accurately recognize and accurately recognize and determine the shape of tables which have rounded or curved corners or accurately recognize and determine the shape of tables which have rounded or curved corners, leading to inaccurate data extraction from non-rectangular or non-square table structures.
Innovation Solution
A system and method for determining the shape of tables in documents using image processing techniques, including image pre-processing, line extraction, boundary detection, cluster detection, and classification, to identify and classify tables with rounded or curved corners.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional table shape determination techniques are used, then rectangular or square-shaped tables can be recognized, but tables with rounded or curved corners cannot be accurately detected
Solution Approach 1:
The patent segments the table detection process into distinct phases: initial rectangular table detection using conventional techniques, identification of corner regions, and classification of corner types (pointed vs. curved). This segmentation allows the system to handle both rectangular and non-rectangular tables by treating corner analysis as a separate, specialized sub-task.
Solution Approach 2:
Instead of attempting to detect all table shapes simultaneously using a complex algorithm, the patent inverts the approach by first detecting tables as rectangular and then analyzing corner characteristics to determine if they deviate from rectangular shapes. This inversion simplifies the overall detection process while improving versatility.
2Device complexity
If only rectangular or square tables are recognized, then the detection process remains simple, but tables with different structures cannot be accurately determined
Solution Approach 1:
The patent performs preliminary detection of tables as rectangular structures using simple conventional techniques. Only after this preliminary step does the system proceed to analyze corner regions for curvature. This preliminary action maintains simplicity for standard cases while enabling accurate detection of non-rectangular tables when needed.
Solution Approach 2:
The patent introduces corner analysis as an intermediary step between initial table detection and final shape classification. By analyzing corner regions separately and using them as mediators to determine overall table shape, the system achieves high reliability without requiring completely complex detection algorithms.
3Measurement precision
If advanced image processing techniques are used to detect rounded corners, then detection accuracy improves, but processing complexity increases
Solution Approach 1:
The patent applies advanced image processing techniques locally only to corner regions rather than to the entire table structure. By focusing computational resources on specific corner areas and using techniques like contour analysis and curvature calculation only where needed, the system achieves high detection precision while minimizing overall processing complexity.
Data Source
Figure 1~2
Figure 3A
Figure 3B~3C
AI summary
A method and system of determining shape of a table in a document is disclosed. A region of interest (ROI) from a binarized image of the document is determined is detected corresponding to the table based on detection of a plurality of lines. The ROI is extracted based on a minimum height threshold and a minimum width threshold of the document image. A cluster of points corresponding to each corner of the ROI are determined based on a height of the ROI and contour detection. A corner type of each corner is determined to be one of a pointed corner or a curved corner and in case the corner type of least two corners is determined as the curved corner the shape of the table is determined as a rounded corner structure.