Adaptive Reference Image Selection for Motion Distortion Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Image fusion techniques often result in motion distortion due to differences in capture time, camera shake, and rotation between images, leading to ghosting artifacts and visible edges in the final fused image, especially in low-light conditions with small pixel devices.
Innovation Solution
An adaptive approach to selecting a reference image from an incoming image stream using a robust motion detection analysis process, which evaluates image frame pairs with differing exposure settings to determine the best image for fusion, reducing motion distortion by choosing between underexposed and normally-exposed images based on motion levels.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple images are fused to increase signal-to-noise ratio, then image quality improves, but motion distortion and ghosting artifacts worsen
Solution Approach 1:
The patent applies preliminary action by performing motion detection and reference image selection before the image fusion process. The system analyzes image frame pairs to detect motion and identifies the best reference image in advance, ensuring that the fusion operation starts with optimally selected images that minimize motion distortion while maximizing signal-to-noise ratio.
Solution Approach 2:
The patent implements dynamics by making the reference image selection adaptive and dynamic rather than static. The system continuously evaluates image frame pairs using motion detection algorithms and adjusts the reference image selection based on real-time motion conditions, allowing the fusion process to adapt to changing scene dynamics and minimize ghosting artifacts.
2Illumination intensity
If images with different exposure settings are used, then low-light image quality improves, but motion distortion increases
Solution Approach 1:
The patent applies local quality by making the reference image selection scene-dependent and region-aware. The system evaluates image frame pairs and selects reference images based on local motion characteristics and exposure quality in different regions of the scene, allowing optimal exposure settings to be utilized while minimizing motion distortion in critical areas.
Solution Approach 2:
The patent implements parameter changes by dynamically adjusting the selection of images with different exposure settings based on detected motion levels. The system changes the exposure parameters of the selected reference image according to scene conditions, utilizing underexposed images when motion is detected and normally-exposed images when the scene is stable, thereby optimizing both low-light quality and motion distortion reduction.
3Reliability
If image registration is performed on images with significant time differences, then more images can be fused for better noise reduction, but registration accuracy deteriorates
Solution Approach 1:
The patent applies preliminary action by performing motion detection and reference image selection before the registration and fusion processes. This preliminary analysis ensures that only images with acceptable registration potential are selected, improving the accuracy of subsequent registration operations while still allowing multiple images to be fused for noise reduction.
Data Source
AI summary
Devices, methods, and computer-readable media describing an adaptive approach to reference image selection are disclosed herein, e.g., to generate fused images with reduced motion distortion. More particularly, an incoming image stream may be obtained from an image capture device, which image stream may comprise a variety of different image captures, e.g., including “image frame pairs” (IFPs) that are captured consecutively, wherein the images in a given IFP are captured with differing exposure settings. When a capture request is received at the image capture device, the image capture device may select two or more images from the incoming image stream for fusion, e.g., including at least one IFP. In some embodiments, one of the images from the at least one IFP will be designated as the reference image for a fusion operation, e.g., based on a robust motion detection analysis process performed on the images of the at least one IFP.