Image-Based Object Repair Assessment and Adhesive Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods fail to accurately identify objects in an environment that require repair and suggest appropriate adhesives for their repair using captured images, especially due to variations in lighting and angles.
Innovation Solution
A method utilizing machine learning models to analyze digital images from different angles and lighting conditions to identify object features, determine repair measures, and select suitable adhesives, integrated with augmented reality for guidance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a single digital image is used to identify objects and determine repair needs, then the processing time is short, but the identification accuracy is insufficient due to variations in lighting and angles
Solution Approach 1:
The system segments the identification task by analyzing multiple digital images (captured from different angles and lighting conditions) separately, then combines their features to form a comprehensive object characterization. This segmentation allows accurate feature extraction from each individual image while collectively achieving high identification accuracy across varying conditions.
Solution Approach 2:
The system transitions from analyzing a single two-dimensional image to processing multiple images that add temporal and angular dimensions. By capturing images at different times, angles, and lighting conditions, the system creates a multi-dimensional feature space that improves identification accuracy without being constrained by the limitations of any single image.
2Measurement precision
If multiple digital images from different angles and lighting conditions are processed, then the identification accuracy improves, but the processing time increases
Solution Approach 1:
The system performs preliminary actions by capturing multiple digital images under varying conditions (different angles, lighting, times) before the actual analysis. These pre-captured images provide comprehensive feature data that enables accurate identification without requiring extensive processing during the analysis phase, as the feature extraction is facilitated by the diverse pre-captured dataset.
Solution Approach 2:
The system merges features extracted from multiple individual images into a unified object representation. By combining complementary information from images captured under different conditions, the system achieves robust and accurate object identification that leverages the strengths of each individual image while compensating for their individual limitations.
3Measurement precision
If detailed surface characterization is performed to determine suitable adhesives, then the adhesive selection accuracy improves, but the analysis complexity increases
Solution Approach 1:
The system applies local quality analysis by characterizing specific surface regions with detailed features (porosity, roughness, material composition) rather than treating the entire surface uniformly. This localized detailed analysis of critical surface areas enables accurate adhesive selection for specific repair locations while avoiding unnecessary complexity in analyzing entire objects.
Solution Approach 2:
The system introduces an intermediary layer of surface feature extraction and characterization that bridges the gap between raw image data and adhesive selection. By computing intermediate surface properties (porosity, roughness, material type) from images, the system creates a simplified representation that facilitates accurate adhesive matching without directly analyzing complex raw image data.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
According to a method for image processing is described, the method comprising determining an object in each digital image of a plurality of digital images containing the object, the plurality of digital images capturing the object at different angles and/or under different lighting conditions, determining one or more features for the determined object and determining, using the determined one or more features and by means of a machine learning model, whether a repair measure is required for the object and, if so, which repair measure is required, wherein for the determined repair measure, one or more adhesives to be used within the scope of the determined repair measure are additionally determined from a set of a plurality of predetermined and stored adhesives.