Construction Drawing Area Detection via Merged Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The process of quantity takeoff from 2D construction drawings is labor-intensive and time-consuming, requiring manual measurement of areas and materials, which is inefficient and prone to errors.
Innovation Solution
A computing platform employs multiple image processing techniques, including semantic segmentation, instance segmentation, and unsupervised processing, to automatically detect areas within 2D construction drawings, merging the results to enhance accuracy and efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual quantity takeoff is performed on 2D construction drawings, then measurement accuracy can be maintained through human judgment, but the process becomes extremely time-consuming and labor-intensive
Solution Approach 1:
The patent replaces the manual mechanical measurement process with an automated image processing system that uses computer algorithms to detect walls, doors, windows, and calculate areas automatically. This substitution eliminates the need for human estimators to manually trace and measure construction drawings, dramatically reducing time consumption while maintaining measurement accuracy through consistent algorithmic application.
Solution Approach 2:
The system enables self-service quantity takeoff by automatically processing construction drawings without human intervention. The image processing algorithms independently identify geometric features, calculate areas, and generate quantity data, allowing the system to serve itself rather than requiring external human labor for each measurement task.
2Productivity
If automated image processing is used to detect areas in construction drawings, then processing speed and productivity are significantly improved, but the complexity of the detection system increases
Solution Approach 1:
The patent segments the area detection task into distinct components: wall detection, door detection, window detection, and area calculation. Each component is handled by specialized image processing algorithms that focus on specific features, making the overall system more manageable and efficient despite the complexity of the complete automation process.
Solution Approach 2:
The image processing system is designed with multi-functionality to handle various construction drawing types and formats. The same core algorithms can process different drawing styles, scales, and representations, reducing the need for multiple specialized systems and managing complexity through a unified approach.
3Reliability
If multiple image processing techniques are combined to merge area detection results, then detection accuracy and reliability are enhanced, but the computational complexity and processing time increase
Solution Approach 1:
The patent merges results from multiple image processing techniques by integrating wall detection, door detection, and window detection algorithms. The system combines these separate detection processes to calculate final area measurements, leveraging the strengths of each technique while maintaining overall system coherence through coordinated processing.
Data Source
AI summary
An example computing platform is configured to (i) receive a two-dimensional (2D) image file comprising a construction drawing, (ii) generate, via semantic segmentation, a first set of polygons corresponding to respective areas of the 2D image file, (iii) generate, via instance segmentation, a second set of polygons corresponding to respective areas of the 2D image file, (iv) generate, via unsupervised image processing, a third set of polygons corresponding to respective areas of the 2D image file, (v) based on (a) overlap between polygons in the first, second, and third sets of polygons and (b) respective confidence scores for each of the overlapping polygons, determine a set of merged polygons corresponding to respective areas of the 2D image file, and (vi) cause a client station to display a visual representation of the 2D image file where each merged polygon is overlaid as a respective selectable region of the 2D image file.