Turbulence Correction for Moving Targets in ATR Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Atmospheric turbulence causes image degradation in video data acquired from aircraft-mounted video image acquisition units, leading to blurring and reduced image quality, which affects target recognition and processing in automatic target recognition (ATR) systems.
Innovation Solution
An AI/ML-based ATR system with a turbulence correction method that uses a moving tile algorithm to detect and correct atmospheric disturbances, optimizing high-frequency scene content and suppressing low-frequency noise to improve target recognition, employing a processor that demosaics, processes, and mosaics video tiles to generate a full field of view turbulence-corrected video stream.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If atmospheric turbulence correction is applied to improve image quality, then image resolution and contrast are enhanced, but processing complexity and computational load increase
Solution Approach 1:
The patent divides the video stream into multiple tiles and processes each tile independently through the turbulence correction pipeline. This segmentation allows parallel processing of different regions, reducing overall computational complexity while maintaining image resolution improvements across the entire field of view
Solution Approach 2:
The system dynamically adjusts processing parameters based on detected turbulence conditions and target motion. The turbulence correction strength, tile processing priority, and resource allocation are adapted in real-time based on scene content and atmospheric conditions, optimizing the balance between image quality and processing load
2Reliability
If turbulence correction processing is applied to correct video frames, then image quality and target recognition are improved, but processing time increases
Solution Approach 1:
The system performs preliminary actions by detecting targets and turbulence characteristics before applying full turbulence correction. AI/ML-based target detection identifies regions of interest, and preliminary turbulence assessment allows the system to focus computational resources on critical areas, reducing overall processing time while maintaining recognition accuracy
Solution Approach 2:
The patent applies turbulence correction selectively rather than uniformly across all frames and regions. Full correction is applied to identified targets and high-priority regions, while other areas receive reduced or no processing, achieving acceptable target recognition accuracy with diminished processing time requirements
3Measurement precision
If the entire field of view is processed for turbulence correction, then image quality is improved across all regions, but computational resources are overwhelmed
Solution Approach 1:
The field of view is divided into multiple tiles that can be processed independently and in parallel. This segmentation enables distribution of computational workload across multiple processing units, preventing resource overload while ensuring image quality improvement across the entire field of view through coordinated tile processing
Solution Approach 2:
The system applies different processing quality levels to different regions based on their importance. Identified targets and high-priority regions receive full turbulence correction with maximum image quality, while background and low-priority areas receive reduced processing, optimizing the distribution of computational resources across the field of view
Data Source
AI summary
A system, and method of operating the same detects moving targets in images and performs image turbulence correction. The system includes an automatic target recognizer (ATR) system including a database. The ATR includes a feature extractor and processor arranged to detect a plurality of reference features associated with targets within image frames, and calculate a position of the plurality of reference features. The system includes an image processor arranged to receive the position, demosaic the image frames into a plurality of video tiles, iteratively process the video tiles for turbulence correction to generate turbulence corrected video tiles associated with acquired targets; convert the turbulence corrected video tiles into a single video frame tile including turbulence degradation correction; and mosaic each of the single video frame tiles to generate a full field of view turbulence corrected video stream.