Composite Image Generation for Wind Turbine Blades
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for generating composite images of elongated objects, such as wind turbine blades, face challenges in robustness and computational efficiency due to the need to combine overlapping images captured from different perspectives, which introduces distortions and increases cognitive load for both human and automated inspection systems.
Innovation Solution
A method involving the retrieval of overlapping source images, normalization, 3D data analysis, feature detection, and image transformation to fuse images while maintaining geometric relations, specifically using edge features aligned with the longitudinal axis to simplify and streamline the process, reducing the complexity of feature matching and improving accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If multiple overlapping images are captured from different perspectives to show the entire surface of an elongated object, then the completeness of surface inspection is improved, but the cognitive load on users and complexity of automated analysis increases due to the need to account for different perspectives and overlapping regions
Solution Approach 1:
The patent segments the elongated object surface into multiple overlapping regions, each captured in a separate source image. This segmentation allows the entire surface to be inspected by combining multiple manageable image sections rather than requiring a single complex overview image, thereby reducing the cognitive load while maintaining inspection completeness
Solution Approach 2:
The patent transforms 2D source images into a 3D representation of the elongated object surface. By adding the dimensional aspect of depth and spatial positioning, the system can accurately represent the surface geometry and eliminate distortions caused by different camera perspectives, making the composite image easier to interpret while maintaining completeness
2Quantity of substance
If multiple images are combined to show the entire object surface, then the completeness of inspection is improved, but the accuracy of feature matching deteriorates due to distortions from different perspectives
Solution Approach 1:
The patent creates a 3D representation of the object surface from multiple 2D source images. This dimensional transformation allows accurate geometric relationships to be maintained, as the 3D model accounts for the spatial positioning and perspective of each source image, thereby improving feature matching accuracy while maintaining inspection completeness
Solution Approach 2:
The patent uses a 3D representation as an intermediary between the multiple source images and the final composite image. This intermediary model serves as a reference framework that aligns features across different perspectives, eliminating distortion effects and improving measurement precision while combining multiple images for complete surface inspection
3Ease of operation
If a composite image combining multiple source images is generated, then the ease of operation for human users is improved by reducing cognitive load, but the computational efficiency deteriorates due to the complexity of image processing required
Solution Approach 1:
The patent segments the image processing into distinct modular steps: retrieving source images, generating normalised images, detecting characteristic features, determining transformations, and forming the composite image. This segmentation allows each step to be optimized independently, improving computational efficiency while producing a user-friendly composite image that reduces cognitive load
Solution Approach 2:
The patent performs preliminary processing steps including generating normalised images and detecting characteristic features before creating the final composite image. These preliminary actions prepare the data in advance, making the subsequent composite image formation more efficient and the final result easier to interpret for human users, thereby improving both productivity and ease of operation
4Measurement precision
If feature detection and image transformation are performed to align images, then the accuracy of composite image generation is improved, but the computational time required increases
Solution Approach 1:
The patent focuses feature detection and image transformation operations on specific local regions containing characteristic features rather than processing the entire image uniformly. This local quality approach concentrates computational resources where they are most needed for accurate feature matching, improving accuracy while reducing overall computational time
Solution Approach 2:
The patent performs feature detection and transformation determination as continuous iterative processes during the image generation workflow rather than as separate discrete steps. This continuous approach integrates the computations efficiently, maintaining accuracy of feature matching while minimizing idle computational time and overall processing duration
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
A composite image of a surface of an elongated object (12), such as a wind turbine blade, is generated from source images (1) by mapping them onto a reference surface (14), thereby generating normalised images (2). Characteristic image features (3) are detected, and pairs of normalised images (2) are fused by determining a relative transformation so that the positions of corresponding characteristic features in the images match. An image transformation is determined that, when applied to the normalised images (2) results in geometric relations between the characteristic features in the normalised images (2) being at least approximately the same. A fused image (5) is formed by applying the image transformation and the relative transformation to the normalised images (2) and combining the resulting images.