Image Restoration With Physical Scattering Model for Non-Uniform Illumination
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image restoration methods fail to effectively address backscattering and light attenuation in scattering media, leading to color distortion and blurring, especially in non-uniform illumination conditions, which affects the performance of computer vision algorithms.
Innovation Solution
A physical scattering model is established to map the relationship between the light source position and scattering, dividing the image into direct radiation and backscattering components, and using parameter fitting to restore a clear image by removing backscattering and attenuation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If the atmospheric scattering model based on Beer-Lambert law is used, then the model simplicity is improved, but the ability to cover complex situations deteriorates
Solution Approach 1:
The patent segments the scattered light into two distinct components: backscattering component and direct radiation component. This segmentation allows the model to handle complex illumination scenarios by treating each component separately with appropriate physical models, thereby improving adaptability while maintaining reasonable complexity
Solution Approach 2:
The patent introduces light source position parameters (azimuth angle and zenith angle) to transform the model from a simple depth-based approach to one that accounts for directional illumination. This parameter change enables the model to adapt to various lighting conditions and complex scattering scenarios
2Reliability
If the atmospheric scattering model is used, then the removal of scattering from air and water is improved, but the problem of color distortion and blurring deteriorates
Solution Approach 1:
The patent extracts the backscattering component from the total scattered light using the derived mapping relationship with light source position. By separating and removing only the backscattering component while preserving direct radiation, the model effectively removes scattering effects without causing color distortion or blurring
Solution Approach 2:
The patent uses the light source position parameters as an intermediary to establish the mapping relationship between scattering characteristics and physical parameters. This intermediary enables accurate estimation of backscattering without directly measuring it, thereby improving scattering removal while avoiding artifacts
3Productivity
If the atmospheric scattering model is used, then the processing speed is improved, but the performance of computer vision algorithms deteriorates
Solution Approach 1:
The patent performs preliminary estimation of light source position parameters from the input image before applying the scattering removal model. This preliminary action provides accurate initial parameters for the model, enabling fast processing while ensuring high-quality results that maintain computer vision algorithm performance
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The method effectively removes backscattering and light attenuation, providing natural image restoration under non-uniform illumination, especially in high-density scattering environments, outperforming existing algorithms in stability and effectiveness.
Implementation Method 1
The well-known atmospheric scattering model is based on the Beer-Lambert law: I(x)=J(x)t(x)+A(1−t(x)), t(x)=e−σz
Implementation Method 2
dividing the attenuated image into a backscatter component and a direct radiation component for calculation
Data Source
AI summary
A new image formation model for descattering different from the previous works based on atmospheric scattering model is established. It can provide a corresponding physical explanation for each scattering environment, illustrate a relationship between a light source position and scattering conditions and solve the problem of scattering removal under non-uniform illumination. Firstly, a dark angle of a camera is removed by an integrating sphere; then, a pure scattering part is found by an improved dark channel-like prior method; and finally, a zenith angle, an azimuth angle, an extinction coefficient and other related physical quantities are calculated according to the light attenuation and backscattering distribution of a scattering medium to eliminate the backscattering and light attenuation. Theoretically, the model can be used in any uniform scattering medium, such as underwater and fog environments, and works well in high-density media.