Airport Scene Perception With Logic-Validated Feature Categories
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing machine learning-based approaches for perceiving airport scenes are not optimal or robust in some situations, leading to inaccuracies and reduced reliability in identifying airport features.
Innovation Solution
A method combining machine learning with logical reasoning to process images of airport scenes, applying predetermined logical tests to initial machine learning estimates to refine and verify the accuracy of airport feature categorization, using multiple imaging modalities and segmentation techniques.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If machine learning algorithms are used to identify airport features in images, then the processing speed and automation level are improved, but the accuracy and reliability of feature categorization deteriorate in some situations
Solution Approach 1:
The system implements feedback by applying predetermined logical tests to the output of machine learning algorithms. The logical reasoning module receives initial estimates from the ML model, validates them against known constraints and relationships, and refines the categorizations based on this feedback loop, thereby improving reliability while maintaining automation.
Solution Approach 2:
The system combines multiple approaches (machine learning algorithms and logical reasoning) into a composite system. Rather than relying on a single method, it integrates the pattern recognition capabilities of ML with the constraint-based validation of logical tests, creating a more robust hybrid approach that leverages the strengths of both methodologies.
2Productivity
If machine learning algorithms are used to identify airport features, then productivity is improved, but measurement precision of feature categorization deteriorates
Solution Approach 1:
The logical reasoning module acts as an intermediary between the machine learning algorithm and the final output. It receives the ML model's initial estimates, applies validation rules and logical constraints, and produces refined categorizations. This intermediary layer maintains processing efficiency while improving measurement precision through systematic validation.
3Device complexity
If only machine learning is used for airport scene perception, then device complexity is reduced, but accuracy of distinguishing airport features from non-airport features deteriorates
Solution Approach 1:
The system merges machine learning-based pattern recognition with logical reasoning-based validation into a unified airport scene perception system. This combination allows the system to leverage both data-driven insights from ML and rule-based certainty from logical tests, improving the accuracy of distinguishing airport features from non-airport features while maintaining reasonable system complexity.
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
A method of perceiving an airport scene is disclosed. The method comprises receiving an image of at least part of an airport and processing (304) said image using a first machine learning algorithm to identify at least one segment (402) in said image and determine an initial estimate of a category of airport feature present in said segment. The method also comprises determining (310) one or more real-world coordinates or dimensions associated with the segment, applying (312) one or more predetermined logical tests to the initial estimate of the category of airport feature present in the segment and the real-world coordinates or dimensions of the segment to determine a reviewed estimate of the category of airport feature present in the segment, and outputting (320) said reviewed estimate.