Turbomachine Surface Imaging via Partial Image Stitching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
There is a need for an effective method to image and detect damage on the surfaces within the internal spaces of axial turbomachines, such as aircraft engines, where direct visualization is challenging due to the complex geometry and accessibility issues.
Innovation Solution
The method involves using an imaging device, such as an endoscope, to capture sequential partial images of the surface from different positions, which are then stitched together to form a comprehensive image. This process allows for the objective evaluation of the surface and detection of damage, utilizing techniques like feature extraction and tracking, and positional data integration for accurate stitching.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If a single image is used to capture the surface, then the imaging process is simple and fast, but the entire surface cannot be captured due to limited field of view
Solution Approach 1:
The surface imaging is divided into multiple sequential partial images captured by the imaging device at different positions. Each partial image covers a specific region of the surface, and these segments are later stitched together to form a complete image of the entire surface, resolving the contradiction between capturing large area and maintaining simple imaging process
Solution Approach 2:
The imaging process transitions from a single two-dimensional image to a sequence of partial images captured at different spatial positions. By adding the temporal dimension of sequential capture and positional information, the system achieves complete surface coverage while maintaining the simplicity of individual image capture operations
2Area of stationary object
If sequential partial images are captured and stitched together, then complete surface coverage is achieved, but the evaluation process becomes more complex
Solution Approach 1:
The surface is divided into multiple partial regions captured in sequential images. Each partial image can be independently evaluated for damage, and the results are aggregated to assess the overall surface condition. This segmentation enables comprehensive damage detection while maintaining manageable complexity through localized analysis
Solution Approach 2:
Multiple copies of the surface at different positions are captured as sequential partial images. These copies are then stitched together to create a comprehensive view, enabling objective evaluation and damage detection across the entire surface while maintaining the simplicity of individual image analysis
3Ease of operation
If the imaging device captures the entire surface at once, then the imaging process is simple, but the device cannot access certain areas due to geometric constraints
Solution Approach 1:
The imaging process is segmented into multiple partial captures at different positions. The imaging device can access each local region independently, and the results are combined to provide comprehensive surface coverage. This resolves the contradiction by enabling operation simplicity at each segment while achieving overall versatility through sequential positioning
Solution Approach 2:
The imaging device dynamically changes its position to capture different partial regions of the surface. By making the imaging process adaptive and flexible rather than static, the system can access various surface areas while maintaining the simplicity of individual capture operations through controlled movement and positioning
Data Source
AI summary
The present invention relates to a method for imaging a surface in an internal space of an axial turbomachine, wherein the surface is captured in a sequence using an imaging device, i.e.—only a portion is captured with the imaging device at a respective time, thus generating only a partial image of the surface; the imaging device and the surface are moved relative to each other such that different portions of the surface are captured over the course of time, thus generating different partial images, and wherein the partial images are stitched together to produce an image of the surface.