HDR Image Generation via Motion-Adaptive Weighted Merging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing imaging technologies struggle to capture high dynamic range images effectively, leading to loss of detail in bright or dark areas, as they often rely on single exposure images or simple merging techniques that fail to account for motion and saturation levels.
Innovation Solution
A method and apparatus for generating a high dynamic range (HDR) image by capturing a long exposure image and a short exposure image, computing merging weights for each pixel based on pixel values and saturation thresholds, and calculating pixel values for the HDR image as a weighted sum of corresponding pixels from the long and short exposure images, while adapting for motion in the scene.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Illumination intensity
If multiple exposures are merged to generate HDR image, then dynamic range is improved, but ghosting artifacts occur due to motion in the scene
Solution Approach 1:
The patent applies local quality by computing different merging weights for different pixel locations based on local motion detection and saturation levels. The merging weight for each pixel is determined by comparing pixel values across exposures and detecting motion locally, allowing the system to preserve high dynamic range benefits in static regions while minimizing ghosting artifacts in regions with motion. This localized approach enables different parts of the image to have different merging strategies.
Solution Approach 2:
The patent changes parameters by dynamically adjusting merging weights based on pixel saturation thresholds and motion detection results. The merging weight varies from 0 to 1 depending on the degree of saturation and detected motion at each pixel location. This parameter adjustment allows the system to adaptively balance between preserving dynamic range and minimizing artifacts based on local image characteristics.
2Device complexity
If simple merging techniques are used, then processing complexity is reduced, but detail preservation in bright or dark areas is lost
Solution Approach 1:
The patent applies preliminary action by performing motion detection and saturation analysis before the merging process. The system detects motion in the scene and identifies saturated pixels in advance, then uses this information to compute appropriate merging weights for each pixel location. This preliminary analysis enables the merging process to preserve details in both bright and dark areas without requiring complex iterative optimization during the actual merging step.
3Productivity
If merging weight is computed based on saturation threshold only, then processing speed is improved, but motion artifacts are not minimized
Solution Approach 1:
The patent applies segmentation by separating the merging weight computation into two independent components: saturation-based weighting and motion-based weighting. The system first computes weights based on saturation thresholds, then separately detects motion and adjusts weights accordingly. This segmentation allows the system to maintain processing efficiency while incorporating motion information to minimize artifacts, as each component can be computed independently and combined.
Data Source
AI summary
A method of generating a high dynamic range (HDR) image is provided that includes capturing a long exposure image and a short exposure image of a scene, computing a merging weight for each pixel location of the long exposure image based on a pixel value of the pixel location and a saturation threshold, and computing a pixel value for each pixel location of the HDR image as a weighted sum of corresponding pixel values in the long exposure image and the short exposure image, wherein a weight applied to a pixel value of the pixel location of the short exposure image and a weight applied to a pixel value of the pixel location in the pixel long exposure image are determined based on the merging weight computed for the pixel location and responsive to motion in a scene of the long exposure image and the short exposure image.