Image Enhancement Control Curve for Brightness and Contrast
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image enhancement methods struggle to simultaneously control brightness, contrast, and noise in video images, often magnifying existing differences or contrast, which affects the preservation of image details.
Innovation Solution
A method involving the acquisition of luminance statistics to determine a boosting curve for target brightness and an enhancing curve based on saliency distribution and noise level, blending these curves to generate a control curve that enhances image brightness and contrast, thereby preserving more details.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If conventional tone mapping and content-based enhancement are used, then image quality is enhanced, but brightness, contrast and noise cannot be controlled simultaneously
Solution Approach 1:
The patent segments the image enhancement process into multiple independent modules: luminance statistics analysis, boosting curve generation, enhancing curve generation, and curve blending. Each module handles specific aspects (brightness, contrast, noise) separately, allowing independent control and adjustment of each parameter without affecting others, thus resolving the contradiction between enhancing image quality and controlling multiple parameters simultaneously.
Solution Approach 2:
The patent changes key parameters by generating adaptive curves (boosting curve and enhancing curve) based on luminance statistics and saliency distribution. These curves transform pixel values to achieve target brightness and contrast while controlling noise. The blending parameter alpha allows continuous adjustment between different enhancement strategies, providing precise control over the enhancement process.
2Illumination intensity
If conventional enhancement methods are applied, then image contrast is enhanced, but existing differences or contrast are magnified
Solution Approach 1:
The patent applies local quality enhancement by generating an enhancing curve based on saliency distribution of pixel brightness. Different regions of the image receive different enhancement strengths according to their local characteristics - important regions (high saliency) are enhanced more while preserving details, and less important regions are enhanced less to avoid magnifying noise and artifacts. This local adaptive approach enhances contrast without magnifying existing differences excessively.
Solution Approach 2:
The patent incorporates feedback mechanisms by analyzing luminance statistics and using them to adaptively adjust the enhancement curves. The boosting curve and enhancing curve are generated based on the actual luminance distribution and saliency characteristics of the input image, creating a feedback loop that ensures contrast enhancement is proportional to the actual image content and does not excessively magnify existing differences or lose important details.
3Manufacturing precision
If brightness and contrast are enhanced, then image quality is improved, but noise control becomes difficult
Solution Approach 1:
The patent performs preliminary analysis of luminance statistics and saliency distribution before applying enhancement. The boosting curve and enhancing curve are pre-computed based on the statistical characteristics of the image, allowing the system to anticipate where enhancement is needed and where noise control is critical. This preliminary action enables the system to enhance quality while preemptively controlling noise in regions where it would be problematic.
Solution Approach 2:
The patent introduces blending curves (boosting curve and enhancing curve) as intermediaries between the original image and the enhanced output. These curves act as mediators that transform pixel values in a controlled manner, allowing the system to achieve both quality enhancement and noise control. The alpha blending parameter provides additional control to balance between different enhancement effects and noise suppression.
Data Source
AI summary
A method of image enhancement in a video, comprises acquiring a luminance statistics of an image; determining a boosting curve according to a target brightness and a brightness of the image; determining an enhancing curve according to a saliency distribution of a pixel brightness derived from the luminance statistics, to the target brightness and a noise level of the image; blending the boosting curve and the enhancing curve based on an enhance level to generate a control curve; and enhancing the image based on the control curve; wherein the control curve enhances the brightness and a contrast of the image.