Image Correlation Assessment via Pixel Partition Luminance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for assessing image correlation in Simulation-Based Training (SBT) within the LVC network architecture are subjective and do not objectively measure rendered images, leading to potential differences in visual fidelity experienced by trainees, which can impact training validity and safety.
Innovation Solution
A method that partitions images into pixel partitions, calculates average luminance values, and determines correlation by comparing these values to establish a percentage of correlated partitions, exceeding a predetermined threshold to assess image correlation objectively and automatically.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If TDB correlation methods are used to assess image correlation, then the assessment process is simplified by using underlying data, but the measurement accuracy deteriorates because proprietary algorithms and hardware differences are excluded from the comparison
Solution Approach 1:
The patent creates a digital copy of the rendered image and processes this copy through automated algorithms rather than comparing original images directly or using underlying TDB data. This allows accurate visual correlation assessment while maintaining operational simplicity through software-based processing.
Solution Approach 2:
The patent replaces manual visual inspection methods with automated computer-based image processing algorithms. The system uses software to automatically compare rendered images, calculate correlation metrics, and generate reports, eliminating the need for human observers while improving measurement accuracy and consistency.
2Extent of automation
If human visual inspection is used to compare generated images, then subjective assessment can be performed, but the automation level deteriorates and resource requirements increase
Solution Approach 1:
The patent implements a self-service automated system where the computer automatically performs image correlation assessment without human intervention. The system loads images, processes them through correlation algorithms, and generates results autonomously, eliminating the need for human inspectors while maintaining high assessment quality.
Solution Approach 2:
The patent substitutes human visual inspection with automated computer-based image processing. The system uses software algorithms to perform correlation analysis, replacing the mechanical process of human observation and comparison with automated digital processing that requires no human resources.
3Reliability
If automated image comparison is implemented, then objectivity and consistency improve, but the device complexity increases due to processing requirements
Solution Approach 1:
The patent divides the image correlation assessment process into distinct processing stages: loading images, partitioning images into zones, comparing zone characteristics, calculating correlation metrics, and generating reports. This segmentation allows each stage to be handled by specialized software modules, improving reliability while managing system complexity through modular design.
Solution Approach 2:
The patent transforms the image correlation assessment from a subjective visual process to an objective quantitative process by changing the parameters being measured. Instead of relying on human perception, the system measures objective parameters such as luminance, color values, and spatial relationships, enabling automated processing with high consistency.
Data Source
AI summary
A system and method for determining if a first image and a second image are correlated images includes partitioning a first image and a second image into a plurality of corresponding pixel partitions, calculating an average luminance value for each of the plurality of pixel partitions, determining if each of the plurality of pixel partitions of the first image is correlated with each of the corresponding plurality of pixel partitions of the second image, calculating a percentage of correlated pixel partitions of the first image and the corresponding plurality of pixel partitions of the second image and determining that the first image and the second image are correlated images if the percentage of correlated pixel partitions exceeds a predetermined pixel partition correlation threshold. The objective metric of the present invention determines whether two static rendered images are correlated enough to be undetectable by a human observer.