Adaptive Contrast Transfer Curves for Luminance Region Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional contrast adjustment methods in displays do not account for input image content, leading to unintended average brightness shifts and saturation or clipping, particularly when increasing contrast in images with good contrast or decreasing contrast in images with poor contrast.
Innovation Solution
An adaptive contrast enhancement method that generates transfer curves to alter pixel luminances, enhancing bright pixels in dark images and dark pixels in bright images, while maintaining luminance in mid- and high-luminance regions to avoid unwanted changes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If traditional non-adaptive contrast adjustment is applied, then contrast enhancement is achieved, but unintended average brightness shifts and saturation or clipping occur
Solution Approach 1:
The patent applies local quality by creating different transfer curves for different luminance regions (dark, mid-tone, bright) based on the histogram distribution. Instead of applying a single global contrast adjustment, the system selectively enhances specific luminance ranges while preserving others, thereby achieving precise contrast enhancement without causing unwanted brightness shifts or saturation in regions that already have appropriate luminance levels.
Solution Approach 2:
The patent dynamically changes the transfer curve parameters based on the input image's histogram characteristics. By analyzing the luminance distribution and selecting or generating appropriate transfer curves, the system adapts the contrast enhancement parameters to match the specific content of each image, preventing over-enhancement and saturation while maintaining precision in the enhancement process.
2Manufacturing precision
If contrast is increased in images with good contrast, then contrast enhancement is applied, but brightness shifts and saturation occur
Solution Approach 1:
The system applies local quality by identifying which luminance regions actually need enhancement based on histogram analysis. For images that already have good contrast, the transfer curve is designed to enhance only specific under-represented luminance ranges while leaving well-balanced regions unchanged, thereby avoiding saturation and clipping in regions that would otherwise be unaffected.
3Manufacturing precision
If contrast is decreased in images with poor contrast, then dark areas are reduced, but images with already poor contrast are blacked out
Solution Approach 1:
The patent applies local quality by selectively adjusting only the luminance ranges that require improvement. When decreasing contrast in bright images, the transfer curve is designed to reduce luminance only in the excessively bright regions while preserving mid-tone and shadow details, thereby avoiding the blacking out effect that would occur with global contrast reduction.
4Manufacturing precision
If adaptive contrast enhancement is applied, then image contrast and clarity are improved, but device complexity increases
Solution Approach 1:
The patent applies segmentation by dividing the luminance range into distinct regions (dark, mid-tone, bright) and creating separate transfer curve segments for each region. This segmentation approach allows the system to achieve adaptive contrast enhancement with relatively simple processing, as each segment can be independently controlled and applied, reducing the overall computational complexity compared to more sophisticated adaptive methods.
Data Source
AI summary
Disclosed is a method for generating transfer curves for adaptive contrast enhancement. Transfer curves are generated so as to enhance the bright pixels in mostly dark images and the dark pixels in mostly bright images. For a dark image, a transfer curve is generated which increases luminance in high-luminance regions of the image without substantially changing the luminance in the mid- and low-luminance regions of the image. For a bright image, a transfer curve is generated which decreases the luminance in the dark areas of the image without substantially changing the luminance in the mid- and high-luminance regions of the image. For a medium-bright image, a transfer curve is generated with increases the luminance in the high-luminance range and decreases the luminance in the low-luminance range of the histogram without substantially changing the luminance in the medium-luminance range of the image.