Engineering Drawing Table and Figure Extraction by Text Density
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods are inefficient in segregating table and figure regions from engineering drawings, which are crucial for quality control, due to the complex intermixing of text, tables, and graphics at various angles.
Innovation Solution
An image processing method involving morphological techniques to detect candidate table regions, calculate textual region density, and use contour-based detection to segregate table and figure regions, employing a system with a processor and memory to execute these processes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional text/graphics separation methods are used on engineering drawings, then text and graphics can be separated, but the method is inefficient and cannot effectively handle complex intermixing of text, tables, and graphics at various angles
Solution Approach 1:
The patent segments the engineering drawing into distinct regions (table regions and figure regions) through a multi-step process: first identifying candidate table regions using line intersection detection, then separating them from figure regions based on textual density analysis. This segmentation approach breaks down the complex separation task into manageable stages, improving efficiency while handling various orientations of text, tables, and graphics
Solution Approach 2:
The patent performs preliminary binarization and line extraction before the actual separation process. By converting the image to binary format and extracting horizontal and vertical lines first, the system prepares the data in advance, making the subsequent table region detection and figure region identification more efficient and effective
2Reliability
If preprocessing methods like binarization are applied to engineering drawings, then the image can be converted to bi-level format, but the process requires additional processing steps before advanced methods can be applied
Solution Approach 1:
The patent performs binarization as a preliminary step to convert the engineering drawing to bi-level format, which is necessary for reliable subsequent processing. This preliminary action ensures quality control accuracy by creating a clean binary image that facilitates accurate text/graphics separation, while the efficient implementation minimizes the time loss associated with preprocessing
3Measurement precision
If candidate table regions are detected using line intersection methods, then table regions can be identified, but false positives may occur requiring additional density calculation steps
Solution Approach 1:
The patent uses feedback through textual density calculation to verify candidate table regions. After detecting candidate regions using line intersection methods, the system calculates the textual density within each candidate region and compares it against a threshold. This feedback mechanism eliminates false positives by confirming that true table regions contain sufficient text, thereby improving detection accuracy while managing processing complexity through selective verification
Data Source
AI summary
A system and method of extracting tables and figures from a drawing document is disclosed. The method may include processing coloured image to segmented binary image and extracting a plurality of horizontal lines and a plurality of vertical lines from a foreground of the image. The method may further include detecting a set of candidate table region from the plurality of horizontal lines and the plurality of vertical lines in the image. Further, the method may include calculating textual region density corresponding to each of the set of candidate table regions in the image. The method may further include identifying at least one relevant table region from the set of candidate table regions in the image and a text free region from the at least one additional region in the image. The method may further include identifying at least one figure region from the dilated text free region.