Adaptive Atmospheric Light Map for Dense Haze Image Dehazing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image dehazing algorithms, such as the He algorithm, face challenges in maintaining contrast and details when dealing with images captured in dense haze weather due to uneven illumination, leading to distortion between dark and bright areas.
Innovation Solution
The method replaces the global atmospheric light value with an adaptive linear atmospheric light map that varies according to the concentration degree of haze, using threshold segmentation, normalization, and rotation of the dark channel and atmospheric light maps to achieve a more even distribution of light, thereby improving image dehazing effects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a selected atmospheric light value is used to process the image, then the processing speed is improved, but the contrast of dark area and details of bright area cannot be maintained simultaneously due to uneven illumination
Solution Approach 1:
The patent applies local quality by replacing the global atmospheric light value with a spatially varying atmospheric light map A(x,y) that adapts to different regions of the image. The dark channel prior is used to estimate atmospheric light distribution locally, allowing each region to have its own optimized atmospheric light characteristics, thus maintaining both dark area contrast and bright area details simultaneously
Solution Approach 2:
The patent segments the image processing into multiple stages: dark channel calculation, atmospheric light estimation, transmittance computation, and final image reconstruction. This segmentation allows each component to be optimized independently, with the atmospheric light map providing region-specific parameters that resolve the contradiction between processing efficiency and image quality
2Reliability
If the He algorithm is used for image dehazing, then the dehazing effect is improved, but image distortion occurs in dense haze conditions due to uneven distribution of atmospheric light
Solution Approach 1:
The patent changes the atmospheric light parameter from a single global value to a spatially varying map A(x,y) with different values at different image coordinates. This parameter change allows the algorithm to adapt to varying haze densities across the image, maintaining structural stability in dense haze regions while preserving the enhanced dehazing effect in clearer regions
Solution Approach 2:
The patent introduces dynamics by making the atmospheric light parameter adaptive rather than static. The atmospheric light map is dynamically estimated from the dark channel of each image, allowing the algorithm to automatically adjust to different haze conditions in different regions, thus maintaining both dehazing effectiveness and structural stability
Data Source
AI summary
A method for image dehazing based on adaptively improved linear global atmospheric light of a dark channel. A haze image in haze weather is first obtained, a variation angle of atmospheric light of the image is obtain through calculating a slope of a connection line between a center and a center of gravity of a binary image of the image, a linear atmospheric light map that varies regularly along a variation direction of the atmospheric light is obtains, a dehazed image is solved through an atmospheric scattering model, and then a processed haze image taken in the haze weather is output.