Tabular Information Extraction via OCR Coordinate Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies face challenges in extracting and converting tabular information from image and digital formats into editable formats for automated processing, data analysis, and cognitive solutions, due to issues with image quality, format consistency, and structural variations in tables.
Innovation Solution
A method involving a processor that determines line, word, and character coordinates using OCR techniques, extracts and reconstructs table rows and columns, and outputs the information in formats like HTML, XML, or JSON, facilitating further processing and analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If OCR techniques are used to extract text from documents, then text recognition capability is improved, but accuracy deteriorates due to image quality issues and format variations
Solution Approach 1:
The patent segments the table extraction process into distinct phases: OCR text recognition, coordinate determination, line extraction, row reconstruction, and column reconstruction. By dividing the complex task into manageable segments, each phase can be optimized independently to improve overall accuracy while maintaining adaptability across different document formats.
Solution Approach 2:
The patent introduces intermediate data structures including coordinate information, line metadata, and structured table representations as mediators between the OCR input and final output. These intermediaries facilitate accurate transformation by providing structured intermediate representations that preserve spatial and hierarchical relationships.
2Manufacturing precision
If multiple coordinate systems are determined for lines, words, and characters, then structural reconstruction accuracy is improved, but processing complexity increases
Solution Approach 1:
The patent determines coordinates for different text elements (lines, words, characters) in separate, sequential steps rather than simultaneously. This segmentation allows each coordinate determination to focus on specific spatial relationships, improving accuracy while managing complexity through staged processing.
Solution Approach 2:
The patent performs preliminary coordinate determination and extraction of line metadata before proceeding to row and column reconstruction. By establishing the coordinate framework and extracting essential metadata in advance, subsequent reconstruction phases can proceed more efficiently with pre-processed information.
3Stability of the object's composition
If table structures are reconstructed from extracted lines, then data organization is improved, but processing time increases
Solution Approach 1:
The patent performs preliminary extraction of line metadata including coordinate information and structural properties before reconstruction. By preparing this information in advance, the actual row and column reconstruction processes can proceed more quickly with pre-organized data, reducing overall processing time while maintaining structural integrity.
Solution Approach 2:
The patent creates structured copies of extracted line information in standardized data formats during the reconstruction process. These structured representations preserve the original table structure while enabling efficient manipulation and output generation, balancing organizational accuracy with processing efficiency.
Data Source
AI summary
A method and a system for extracting information from a table in a document is provided. The method includes: receiving a document that includes information that is arranged in a table; determining three sets of coordinates that respectively relate to lines, words, and characters included in the document; extracting a list of lines based on the first set of coordinates; reconstructing the rows of the table based on list of lines and the second set of coordinates; reconstructing the columns of the table based on the reconstructed rows and the third set of coordinates; and outputting a reconstruction of the table. The three sets of coordinates are expressible in an hOCR format that is based on an open standard for representation of scanned information that is obtainable by using an optical character recognition (OCR) technique.