Auto-Contrast Enhancement via HVS Local Difference Histogram Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional contrast enhancement algorithms often result in over-enhancement, noise amplification, and unnatural image appearance due to their inability to distinguish intensity variations in saturated regions, leading to unsuitable enhancement levels and side-effects.
Innovation Solution
An auto-contrast enhancement system utilizing a human visual system-based local difference histogram unit and histogram classifier, which segments intensity values into regions based on sensitivity and applies histogram equalization with side-effect reduction to maintain natural image appearance and suppress noise.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Illumination intensity
If traditional histogram enhancement is applied to improve contrast, then the overall contrast is improved, but over-enhancement and noise amplification occur in saturated regions
Solution Approach 1:
The patent divides the image into different regions based on saturation levels. Saturated regions are identified and separated from non-saturated regions, allowing different enhancement strategies to be applied to each region. This segmentation prevents noise amplification in saturated areas while maintaining contrast improvement in non-saturated areas.
Solution Approach 2:
The patent applies local contrast enhancement only to non-saturated regions where it is beneficial, while avoiding enhancement in saturated regions where it would cause noise amplification and unnatural appearance. This local quality approach ensures that enhancement is applied selectively based on regional characteristics.
2Illumination intensity
If contrast enhancement is applied directly to saturated regions, then intensity variation is enhanced, but the image looks unnatural and noise is amplified
Solution Approach 1:
The patent applies different enhancement qualities to different regions: non-saturated regions receive full contrast enhancement to improve intensity variation, while saturated regions receive limited or no enhancement to preserve natural appearance and avoid noise amplification.
Solution Approach 2:
The patent identifies saturated regions before applying contrast enhancement and takes preliminary action to protect them from over-enhancement. By detecting saturation levels in advance, the system prevents the harmful effects of direct contrast enhancement on saturated regions.
3Illumination intensity
If histogram equalization is applied to achieve uniform intensity distribution, then contrast is improved, but computational complexity increases
Solution Approach 1:
The patent segments the histogram processing into two parts: global histogram equalization for overall contrast improvement and local saturation-based adjustment for regional refinement. This segmented approach achieves uniform intensity distribution while keeping computational complexity manageable through efficient implementation.
4Illumination intensity
If enhancement level is increased to improve contrast, then contrast improvement is enhanced, but side-effects such as over-enhancement increase
Solution Approach 1:
The patent applies different enhancement levels to different regions based on saturation characteristics. Non-saturated regions receive higher enhancement levels for improved contrast, while saturated regions receive lower or zero enhancement to avoid over-enhancement and its associated side-effects.
Solution Approach 2:
The patent applies enhancement selectively only where needed (non-saturated regions) rather than uniformly across the entire image. This partial action approach achieves sufficient contrast improvement without the excessive enhancement that causes side-effects in saturated regions.
Data Source
AI summary
An auto-contrast enhancement system includes a human visual system (HVS)-based local difference (LD) histogram unit configured to build a LD histogram with respect to intensity values; a histogram classifier configured to categorize histograms of input images based on distribution properties; and a histogram equalization (HE) unit configured to process the input image according to a result of the HVS-based LD histogram unit and the enhancement level determined in the histogram classifier.