Image Contrast Enhancement via Local Histogram and Gradient Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional image contrast enhancement methods, such as histogram equalization, fail to effectively improve image clarity in conditions like heavy fog, clouds, dust, or smoke, resulting in unnatural contrast ratios.
Innovation Solution
An image contrast enhancement method and apparatus that estimate the degree of low contrast for each pixel, convert it into an adjustment weight, and calculate brightness and adjustment values to generate histograms, which are then used to adjust pixel values, enhancing contrast and removing clarity-influencing phenomena like fog, dust, or smoke.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Illumination intensity
If histogram equalization is used to enhance image contrast, then the contrast of the image is improved, but the image appears unnatural and the contrast ratio deteriorates in conditions like heavy fog, clouds, dust, or smoke
Solution Approach 1:
The patent applies different processing strategies to different regions of the image based on local characteristics. It identifies sky regions versus non-sky regions and applies histogram equalization selectively, while using gradient-based enhancement for sky regions to maintain natural appearance and avoid unnatural contrast ratios.
Solution Approach 2:
The patent transforms the input image from RGB color space to LAB color space, separating luminance (L channel) from color information (A and B channels). This parameter transformation allows independent processing of brightness and color, enabling contrast enhancement while preserving natural color relationships and avoiding the unnatural effects of direct RGB histogram equalization.
2Illumination intensity
If traditional histogram equalization is applied, then contrast enhancement is achieved, but clarity is reduced due to block artifacts and unnatural rendering in atmospheric conditions
Solution Approach 1:
The patent processes sky regions and non-sky regions differently to preserve clarity. For sky regions, it uses gradient-based enhancement that maintains smooth transitions and avoids block artifacts. For non-sky regions, it applies selective histogram equalization that preserves edge details and prevents the blocky appearance associated with traditional methods.
Solution Approach 2:
The patent introduces gradient calculation as an intermediary step between input and output image processing. By computing gradients and using them to guide the enhancement process, it maintains spatial continuity and prevents the discrete block artifacts that degrade image clarity in traditional histogram equalization methods.
Data Source
AI summary
An image contrast enhancement method and an apparatus thereof are disclosed, which calculate the degree of influencing the clarity according to the influence feature (e.g., heavy fog, dust, smoke, or etc.) in the image, and then adjust the brightness of the pixels corresponding to features of influencing the clarity according to the degree, thereby enhancing image contrast and removing phenomenon of influencing the clarity in the image.