Hand-drawn Table Cell Validation via Edge Orientation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Hand-drawn tables in images pose challenges for image processing devices due to irregular lines, making it difficult to determine table geometry and generate a high-level representation for electronic documents.
Innovation Solution
A method and system that process images by obtaining edges from a skeleton graph, determining orientations, and building cell borders to validate table cells, allowing for accurate detection and representation of hand-drawn tables within electronic documents.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If hand-drawn lines are used to create tables in images, then the tables can be created freely and naturally, but the lines become irregular and curved making it difficult to determine table geometry and validate cell structures
Solution Approach 1:
The patent segments the hand-drawn table into individual cell candidates and further divides each cell into four border segments (top, bottom, left, right). This segmentation allows the system to handle irregular hand-drawn lines by processing each segment independently and validating them against geometric constraints, thereby resolving the contradiction between drawing freedom and geometry detection accuracy.
Solution Approach 2:
The patent transforms the validation approach by changing parameters from checking entire cell contours to validating individual border segments against orientation constraints. Each border segment is validated to ensure it is substantially perpendicular to adjacent borders, providing a robust method that works with irregular hand-drawn lines while maintaining geometric accuracy.
2Device complexity
If traditional edge detection methods are used on hand-drawn tables, then the process is simple, but the irregular lines make it difficult to determine cell positions and orientations accurately
Solution Approach 1:
The patent segments edge detection into obtaining individual edges associated with cell candidates, then further segments each cell into four border orientations. This multi-level segmentation enables accurate position determination by validating each segment independently rather than attempting to process entire irregular cell contours at once.
Solution Approach 2:
The patent adds an orientation dimension to the validation process by determining and validating the orientation of each border segment. Instead of only checking spatial positions, the system validates orientations (e.g., top border should be horizontal, side borders vertical), creating a two-dimensional validation space that improves position determination accuracy.
3Reliability
If the system validates all cell borders to ensure accurate table representation, then the reliability of table detection improves, but the computational complexity and processing time increase
Solution Approach 1:
The patent divides the validation process into four independent border validations (top, bottom, left, right) instead of validating the entire cell at once. Each border is validated against specific orientation constraints, which simplifies the computational complexity while maintaining high reliability through systematic verification of each segment.
Solution Approach 2:
The patent changes the validation parameters from complex geometric shape matching to simpler orientation checks. Each border segment is validated to ensure it is substantially perpendicular to adjacent borders, which is a computationally simpler parameter to check while still providing robust validation reliability.
Data Source
AI summary
A method for processing an image that includes a table is provided. The method includes: obtaining a set of edges associated with a cell candidate of the table; determining a longest edge in the set of edges and a first orientation of the longest edge; determining, in the set of edges, an initial edge for a first cell border comprising the first orientation and the longest edge; building the first cell border based on the initial edge, the first orientation, and the set of edges; calculating a second orientation for a second cell border by rotating the first orientation; building the second cell border based on an initial edge for the second cell border and the second orientation; and validating the cell candidate as a cell of the table in response to building the first cell border and building the second cell border.