Depth Estimation in Distorting Media via Pixel Blurriness
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Underwater and other distorting media images suffer from poor contrast and detail loss due to light absorption and scattering, with existing methods like the dark channel prior failing under varying lighting conditions and exceptions.
Innovation Solution
A method for depth estimation in distorting media using a pixel blurriness map, where a rough depth map is calculated and refined through multi-scale Gaussian-filtering and morphological reconstruction, focusing on background light estimation to adaptively produce accurate scene depth and transmission maps.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If dark channel prior is used to estimate depth and remove haze, then depth estimation is provided, but image contrast remains poor and restoration accuracy deteriorates under varying lighting conditions
Solution Approach 1:
The patent changes the parameter used for depth estimation from dark channel intensity to pixel blurriness. By calculating the difference between original and multi-scale Gaussian-filtered images, the method obtains a blurriness map that reflects actual scene depth more accurately, resolving the contradiction between depth estimation capability and restoration accuracy under varying lighting conditions
2Device complexity
If only blue and green channels are used for depth estimation, then computational complexity is reduced, but restoration quality deteriorates due to exceptions in light absorption and lighting conditions
Solution Approach 1:
Instead of using multiple color channels directly, the patent creates a derived blurriness map through image processing operations (Gaussian filtering and difference calculation). This copied representation of depth information captures essential depth variations without requiring complex multi-channel analysis, achieving both simplicity and high restoration quality
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
This approach enhances image reconstruction by effectively estimating scene depth and transmission maps, improving image quality in various lighting conditions and hues, outperforming previous methods in accuracy and computational efficiency.
Implementation Method 1
Preferred embodiments determined the blurriness map by calculating a difference between an original image and multi-scale Gaussian-filtered images to estimate the pixel blurriness map
Data Source
AI summary
The invention provides a method for depth estimation in image or video obtained from a distorting medium. In the method, a pixel blurriness map is calculated. A rough depth map is then determined from the pixel blurriness map while assuming depth in a small local patch is uniform. The rough depth map is then refined. The method can be implemented in imaging systems, such as cameras or imaging processing computers, and the distorting medium can be an underwater medium, haze, fog, low lighting, or a sandstorm, for example. Preferred embodiments determined the blurriness map by calculating a difference between an original image and multi-scale Gaussian-filtered images to estimate the pixel blurriness map.