Document Table Corner Analysis for Rounded Shape Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional OCR methods struggle to accurately recognize and extract data from tables with non-rectangular shapes, such as those with rounded or curved corners, leading to inaccurate determination of tabular structures.
Innovation Solution
A method and system utilizing image processing techniques to detect regions of interest in documents, including morphological operations, boundary line detection, and contour analysis to classify table shapes based on corner types, distinguishing between pointed and curved corners.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional OCR methods are used to recognize tables, then rectangular or square shaped tables with sharp edges can be recognized, but tables with non-rectangular shapes or curved corners cannot be accurately recognized
Solution Approach 1:
The patent segments the table detection process into distinct phases: initial rectangular table detection using conventional methods, followed by corner analysis by dividing corners into regions and analyzing pixel intensity patterns. This segmentation allows the system to handle both rectangular and curved-corner tables through different analytical approaches.
Solution Approach 2:
The patent applies partial action by focusing analysis only on corner regions of detected tables. Instead of analyzing the entire table, it extracts corner portions and applies specific intensity pattern analysis to these regions, enabling efficient detection of curved corners without processing the whole table structure.
2Productivity
If only rectangular table detection algorithms are used, then simple table structures can be processed quickly, but complex table structures with curved corners require more sophisticated and time-consuming analysis
Solution Approach 1:
The patent performs preliminary detection of rectangular tables using fast conventional algorithms before applying the more complex curved corner analysis. This preliminary action filters out simple rectangular tables that don't require further analysis, applying the sophisticated corner analysis only to candidates that might have curved corners.
Solution Approach 2:
The patent applies different analysis qualities to different parts of the table: fast rectangular detection for the overall table structure, and more intensive pixel intensity pattern analysis specifically at corner regions. This local quality approach optimizes processing speed for rectangular areas while ensuring accuracy at critical corner regions.
Data Source
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 10 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.