Ghost Artifact Removal in HDR Image Fusion via Brightness Deviation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
High Dynamic Range (HDR) image fusion methods struggle with ghost artifacts when objects move or change between different exposure frames, leading to loss of detail and image quality.
Innovation Solution
An image fusion method and device that calculates a ghost weight based on the deviation of brightness parameter ratios between input and reference frames, adjusting the fusion weight to minimize ghost artifacts by using a graphics processing unit to fuse pixels and perform post-filtering processes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Illumination intensity
If multiple image frames at different exposure levels are combined to create HDR images, then the luminance range is expanded and detail in both bright and dark areas is preserved, but ghosting artifacts appear when objects move or change between frames
Solution Approach 1:
The patent changes the fusion weight parameter dynamically based on brightness parameter ratio deviation. By calculating the ghost weight from the deviation amount and using it to adjust the fusion weight, the system adapts to different scene conditions and removes ghosting artifacts while preserving the expanded luminance range achieved through multi-exposure fusion
Solution Approach 2:
The patent implements a feedback mechanism where the brightness parameter ratio is calculated and compared against a reasonable range. The deviation amount feeds back into the ghost weight calculation, which then adjusts the fusion weight. This closed-loop control continuously optimizes the fusion process to eliminate ghosting artifacts while maintaining HDR image quality
2Productivity
If standard fusion algorithms are used to combine image frames, then processing is simple and fast, but ghosting artifacts from moving or changing objects appear in the fused image
Solution Approach 1:
The patent modifies the fusion weight parameter by introducing a ghost weight component based on brightness parameter ratio deviation. This parameter change enables the algorithm to distinguish between genuine scene changes and ghosting artifacts, allowing simple and fast processing while effectively removing ghosting artifacts through adaptive weight adjustment
3Manufacturing precision
If fusion weight is increased to emphasize input frame details, then detail preservation improves, but ghosting artifacts from moving objects are amplified
Solution Approach 1:
The patent applies local quality by making the fusion weight spatially and conditionally variable. Instead of using a uniform fusion weight across the entire image, the system calculates the ghost weight based on local brightness parameter ratio deviation at each pixel location. This allows detail preservation in stable regions while suppressing ghosting artifacts in regions with significant brightness deviations
Solution Approach 2:
The patent dynamically adjusts the fusion weight parameter based on the ghost weight derived from brightness parameter ratio deviation. By changing the fusion weight from a fixed value to a conditionally variable parameter, the system achieves both detail preservation and ghosting artifact removal simultaneously
Data Source
AI summary
An image fusion method for removing ghost artifacts is provided. The method includes: determining a first fusion weight of a reference pixel in a reference frame, wherein the coordinates of the reference pixel are the same as the coordinates of one of input pixels included in an input frame; determining a reference brightness parameter according to values of all channels of the reference pixel; determining an input brightness parameter according to values of all channels of the input pixel; determining a ghost weight according to an amount that the brightness parameter ratio of the input brightness parameter and the reference brightness parameter deviates from a reasonable range; determining a fusion weight according to the first fusion weight and the ghost weight; and fusing the reference pixel with the input pixel according to the fusion weight to generate a fused pixel in a fused frame.