Progressive Image Fusion for Low Noise HDR Capture
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image fusion techniques struggle to generate low noise and high dynamic range (HDR) images in various capturing conditions, particularly in low-light situations, due to predetermined exposure settings and memory limitations in devices like mobile phones, which can result in ghosting artifacts and inadequate signal-to-noise ratio.
Innovation Solution
An adaptive approach to image bracket determination and a memory-efficient image fusion method that analyzes incoming preview images to dynamically adjust exposure times and capture parameters, allowing for progressive fusion of images in batches to maintain low noise and HDR quality across diverse conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple images are captured and fused to reduce noise, then signal-to-noise ratio is improved, but device memory requirements increase
Solution Approach 1:
The patent segments the image fusion process into batches, where images are processed in groups rather than all at once. This allows memory usage to be controlled by limiting the number of images held in memory simultaneously, while still achieving noise reduction through fusion of multiple images across different batches.
Solution Approach 2:
The patent performs preliminary actions by capturing and storing only essential image data and parameters before fusion. By pre-processing images to extract only necessary information and using progressive fusion strategies, memory requirements are reduced while maintaining the ability to achieve high signal-to-noise ratio through multi-image fusion.
2Device complexity
If predetermined exposure settings are used for image capture, then device complexity is reduced, but image quality in varying lighting conditions deteriorates
Solution Approach 1:
The patent implements dynamic exposure control where exposure settings are adjusted based on scene analysis and progression through the capture sequence. Rather than using fixed predetermined settings, the system adapts exposure parameters dynamically to match varying lighting conditions while maintaining manageable device complexity through automated control algorithms.
Solution Approach 2:
The patent changes exposure parameters adaptively during the capture process based on analyzed scene characteristics. By modifying exposure time, gain, and other parameters according to actual lighting conditions detected in different scenes, the system achieves high image quality across diverse conditions without requiring overly complex manual control mechanisms.
3Measurement precision
If images are captured with longer exposure times to reduce noise, then signal-to-noise ratio is improved, but motion artifacts increase
Solution Approach 1:
The patent segments the capture process into multiple exposures of varying durations. By capturing several shorter exposures and fusing them progressively, the system achieves the noise reduction benefits of longer total exposure time while avoiding motion artifacts that would result from any single long exposure. The segmentation allows selective fusion of only those frames with acceptable motion characteristics.
4Manufacturing precision
If adaptive exposure determination is implemented, then image quality in diverse conditions is improved, but processing time increases
Solution Approach 1:
The patent performs preliminary scene analysis and exposure determination based on preview images or initial frame assessment. By determining appropriate exposure parameters in advance before the main capture sequence, the system achieves adaptive image quality optimization without requiring extensive real-time processing during the actual capture, thus minimizing additional processing time.
Data Source
AI summary
An adaptive approach to image bracket determination and a more memory-efficient approach to image fusion, which are designed to generate low noise and high dynamic range (HDR) images in a wide variety of capturing conditions, are described. An incoming preview image stream may be obtained from an image capture device. When a capture request is received, an analysis may be performed on an image from the preview image stream that has a predetermined temporal relationship to the image capture request. Based on the analysis, a set of images (and their respective capture parameters, e.g., exposure time) may be determined for the image capture device to capture. As the determined set of images are captured, they may be registered and fused in a memory-efficient manner that, e.g., places an upper limit on the overall memory footprint of the registration and fusion operations—regardless of how many images are captured in total.