Runway Localization via Image Feature Mirroring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for runway localization in aircraft require high computational effort and specific model knowledge, which is complex, costly, and prone to inaccuracies due to environmental variations, limiting their reliability and applicability.
Innovation Solution
The method involves comparing image features with their mirrored versions to determine the runway center line, leveraging the mirror-symmetrical arrangement of dominant visual components, reducing computational demand and enhancing robustness to visual variations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If feature-based methods or template-based methods are used to localize runways by comparing features with model knowledge, then runway position can be determined, but high computational effort is required in the aircraft
Solution Approach 1:
The patent creates a mirrored copy of the image features and compares the original features with their mirrored versions. This copying approach eliminates the need for external model knowledge databases while maintaining localization accuracy, as the symmetry relationship itself serves as the reference model.
Solution Approach 2:
The method uses the image's own symmetry properties to determine runway position without requiring external model knowledge. The system is self-sufficient, using the mirrored version of its own input data as the reference for comparison, thereby eliminating the need for pre-stored runway models.
2Reliability
If model knowledge about visual components is stored and continuously updated to account for environmental variations, then recognition accuracy can be maintained, but device complexity and cost increase
Solution Approach 1:
Instead of maintaining complex model knowledge databases that require continuous updates for different environmental conditions, the patent creates a simple mirrored copy of the image features. This copying mechanism is universally applicable across all conditions without requiring customization or maintenance of external models.
Solution Approach 2:
The mirrored feature comparison method serves as a universal solution that works across all environmental conditions (daylight, nighttime, various weather) without requiring different models. The same symmetry-based approach is universally applicable, eliminating the need for environment-specific model selection and management.
3Measurement precision
If feature comparison with model knowledge is performed to account for environmental variations like rain, fog, or cloud shadows, then accuracy can be maintained, but computational effort and time increase
Solution Approach 1:
The patent uses a simple mirrored copy operation instead of complex model matching procedures. This copying approach requires minimal computational resources and can be performed rapidly, reducing calculation time while maintaining accuracy through the inherent robustness of symmetry-based comparison.
Solution Approach 2:
Instead of comparing image features against pre-stored models (traditional approach), the patent inverts the approach by mirroring the image features themselves and comparing them against each other. This inversion eliminates the need for time-consuming model selection and matching processes.
4Adaptability or versatility
If extensive model knowledge is required for runway localization, then comprehensive coverage of different runway conditions is achieved, but the system becomes less adaptable to runways without available model knowledge
Solution Approach 1:
The mirrored feature comparison method provides universal adaptability to all runway types and conditions without requiring pre-existing model knowledge. The same symmetry-based approach works for all runways regardless of their specific characteristics, making the system universally applicable.
Solution Approach 2:
The system is completely self-sufficient and does not rely on external model knowledge databases. By using its own mirrored features as the reference, the system can adapt to any runway configuration without requiring pre-stored models or updates, achieving universal coverage through self-service.
Data Source
Figure 1~3
Figure 4a~5a
Figure 5b~6
AI summary
A method and a device for runway localization on the basis of a feature analysis of at least one image of runway surroundings taken by a landing aircraft is characterized in that, in order to determine the central axis of the runway, feature matching between image features (301) of the image and mirrored image features (302) of the image is carried out, wherein features of the one runway side are made to be congruent with features of the other runway side.