AI Corrosion Segmentation for 2D Surface Area Measurement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for calculating corrosion area are time-consuming, costly, and inaccurate, particularly in offshore environments, due to complex geometric patterns, subjective human estimation, and limitations of laser scanners, leading to inefficient and inconsistent measurements.
Innovation Solution
A system comprising modules for image processing using intrinsic and extrinsic calibration, corrosion segmentation, and surface class segmentation, combined with artificial intelligence to calculate the percentage of corroded area from 2D images or video frames, generating textured 3D models.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If laser scanners are used to measure surface area, then measurement accuracy is improved, but device complexity and operating cost increase
Solution Approach 1:
The patent uses 2D images as a simplified copy or representation of the 3D surface, processing these images to extract corrosion area information without requiring complex 3D scanning equipment. This approach maintains measurement capability while significantly reducing device complexity and cost.
Solution Approach 2:
The patent replaces the mechanical laser scanning system with an optical imaging system combined with image processing algorithms. This substitution uses 2D optical images and computational methods to achieve corrosion measurement, eliminating the need for complex mechanical 3D scanning hardware.
2Measurement precision
If 3D engineering models are used to obtain total area, then area measurement is improved, but time consumption and man-hours increase
Solution Approach 1:
The patent extracts only the necessary corrosion area information directly from 2D images without requiring complete 3D modeling of entire structures. By focusing extraction on corrosion-specific features from simplified 2D representations, it eliminates time-consuming comprehensive 3D model creation while maintaining measurement accuracy for the critical parameter.
Solution Approach 2:
The patent segments the image processing task to directly identify and measure corrosion areas without performing complete 3D reconstruction. This selective segmentation approach extracts only the relevant corrosion information from 2D images, bypassing the need for time-intensive full 3D modeling processes.
3Ease of manufacture
If manual measurement methods are used, then device cost is reduced, but measurement accuracy and consistency deteriorate
Solution Approach 1:
The patent implements an automated image processing system that performs corrosion measurement independently without requiring manual intervention. The system automatically processes 2D images, identifies corrosion areas, and calculates measurements, eliminating human subjectivity and inconsistency while maintaining low device costs through software-based automation rather than expensive manual tools.
Solution Approach 2:
The patent replaces manual measurement processes with automated optical image processing. By using 2D image analysis algorithms to automatically detect and measure corrosion, the system eliminates the need for costly manual measurement tools while achieving consistent, accurate results through computational methods.
4Measurement precision
If more images are captured during inspection campaign, then measurement completeness is improved, but capture productivity decreases
Solution Approach 1:
The patent transitions from 3D scanning approaches to 2D image processing, fundamentally changing the dimensional approach to corrosion measurement. This dimensional shift allows for more efficient data capture using standard 2D imaging equipment, improving productivity while maintaining measurement completeness through advanced image analysis that extracts full corrosion information from 2D representations.
Data Source
AI summary
A method for determining the percentage of corroded area from an image, which may be a photograph or a frame from a video. The system includes four main modules: M1—Surface geometry estimation module; M2—Corrosion segmentation module that performs image segmentation to identify corroded and non-corroded surfaces; M3—Optional surface class segmentation module that performs image segmentation to group surfaces by industrial object classes or by specific objects; and M4—Module for calculating the percentage of corroded area.