Atmospheric Light Extraction for Real-Time Image Dehazing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing techniques for removing the influence of fine particles in the atmosphere, such as fog, require significant processing load and are not suitable for real-time applications, leading to reduced visual recognizability of captured images.
Innovation Solution
An image processing apparatus that extracts and removes atmospheric light components using exposure values, separating Mie-scattered and Rayleigh-scattered components, and recombines them to improve image quality in a way that reduces processing load while maintaining real-time capabilities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If histogram processing and multiple threshold value processing are performed across the entire image to estimate atmospheric light component, then the accuracy of atmospheric light estimation is improved, but the processing load becomes heavy and real-time processing is compromised
Solution Approach 1:
The patent extracts atmospheric light component estimation from global image processing and localizes it to specific regions (sky regions or regions with high probability of atmospheric light). By using region-of-interest-based processing and probabilistic mapping approaches, the system estimates atmospheric light only where necessary rather than processing the entire image, thereby reducing computational load while maintaining estimation accuracy.
Solution Approach 2:
The patent segments the image processing task by dividing it into different processing paths: one for sky regions and another for non-sky regions. This segmentation allows the system to apply different processing strategies to different parts of the image, reducing the overall processing load while maintaining accurate atmospheric light estimation in critical areas.
2Productivity
If dedicated circuits are extended for atmospheric light component calculation, then real-time processing capability is improved, but the device complexity and cost increase
Solution Approach 1:
The patent replaces complex dedicated hardware circuits with software-based image processing algorithms that can be executed on general-purpose processors. By using probabilistic mapping methods and region-based processing strategies, the system achieves real-time processing capabilities through efficient algorithm design rather than hardware complexity, thereby reducing device complexity while maintaining real-time performance.
3Reliability
If atmospheric light component is removed from captured image, then visual recognizability is improved, but image processing time increases
Solution Approach 1:
The patent performs preliminary estimation of atmospheric light components and creates probabilistic maps before executing the full dehazing process. By pre-identifying regions with high atmospheric light content and preparing region-specific processing parameters in advance, the system reduces the time required for the actual atmospheric light removal while ensuring visual recognizability is improved in the most critical areas.
Data Source
AI summary
An attempt is made to suppress the processing load while securing real-time properties of the image processing to improve visual recognizability of a captured image whose visual recognizability has been reduced by the influence of fine particle components. An image processing apparatus including: an extraction unit configured to extract an atmospheric light component from a captured image including an influence of fine particles in the atmosphere; and a removal processing unit configured to generate an image from the captured image, in which the influence of fine particles has been removed, based on the extracted atmospheric light component, and the extraction unit performs the extraction based on data of the captured image and data of an exposure value at the time of photographing the captured image.