Image Frame Integration Using Sharpness, Noise, and Jitter Metrics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image frame integration techniques are often degraded by motion blur and noise, failing to effectively combine images due to variations in observation angle and time, leading to suboptimal image quality.
Innovation Solution
A system and method that determine sharpness, noise, and jitter metrics for each image frame, prioritizing or selecting frames based on these metrics to generate an integrated image frame, minimizing the impact of degraded frames and enhancing image resolution.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If image frames are integrated to reduce noise, then noise reduction is improved, but motion blur degrades the integrated image quality
Solution Approach 1:
The patent changes the parameter of exposure time for different image frames, using longer exposure for some frames and shorter exposure for others. This allows selective integration of frames with different noise and motion blur characteristics, optimizing the trade-off between noise reduction and motion blur in the final integrated image
Solution Approach 2:
The patent applies partial integration by selectively choosing which image frames to integrate based on their quality metrics. Not all captured frames are integrated - only those that meet certain criteria regarding noise levels and motion blur, thus avoiding the degradation that would result from integrating all frames equally
2Quantity of substance
If multiple image frames are captured at different time samples, then more data is available for integration, but spatial shifts and jitter degrade image alignment
Solution Approach 1:
The patent performs preliminary registration and alignment of image frames before integration. By pre-correcting spatial shifts and jitter through registration circuits, the system ensures that frames are properly aligned prior to the integration process, maintaining measurement precision even when multiple frames captured at different times are combined
Solution Approach 2:
The patent introduces registration circuits as an intermediary step between image capture and integration. These circuits act as a mediator that corrects spatial misalignments and jitter, enabling the successful integration of multiple frames that would otherwise be degraded by positional variations
3Device complexity
If all image frames are integrated equally, then processing is simplified, but frames with poor quality degrade the overall image
Solution Approach 1:
The patent applies different integration weights to different image frames based on their local quality characteristics. Frames with better sharpness, lower noise, and less motion blur receive higher weights in the integration process, while poorer quality frames receive lower weights. This selective weighting maintains image quality without requiring complex rejection criteria
Solution Approach 2:
The patent uses feedback mechanisms where quality metrics (sharpness, noise, motion blur) are calculated for each frame, and this information feeds back into the integration process to determine optimal weighting. This closed-loop approach automatically adjusts the contribution of each frame based on its actual quality, simplifying the decision process while maintaining high image quality
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The technology described herein includes integration of image frames. The integration of image frames includes determining a sharpness metric for each image frame in a plurality of image frames. The sharpness metric is indicative of at least one of edge content and an edge size of the image frame. The integration of image frames further includes determining a noise metric for each image frame in the plurality of image frames. The noise metric is indicative of a variation in brightness or color in the image frame. The integration of image frames further includes determining a jitter metric for each image frame in the plurality of image frames. The jitter metric is indicative of spatial shifts between the image frame and other image frames in the plurality of image frames. The integration of image frames further includes generating an integrated image frame from one or more of the plurality of image frames based on the sharpness metric, the noise metric, and the jitter metric.