Visual Sub-band Decomposition for Adaptive Image Contrast
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image processing methods fail to effectively address local variations in image contrast and edge detection, leading to suboptimal results in areas with varying illumination, as they often rely on global algorithms that do not adapt well to different regions of an image.
Innovation Solution
The method employs visual sub-band decomposition based on Human Visual System Characteristics (HVSC) to generate multiple sub-band images, which are then processed independently and fused back together, allowing for adaptive image enhancement and edge detection that accounts for local features and varying illumination levels.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If global algorithms are used for image processing, then processing simplicity is maintained, but local variations in contrast and edge detection performance deteriorate
Solution Approach 1:
The image is divided into multiple sub-bands based on human visual system characteristics, with different regions (luminance and chrominance) processed separately. This segmentation allows local adaptation while maintaining overall processing coherence, resolving the contradiction between global simplicity and local precision.
Solution Approach 2:
Different processing strategies are applied to different sub-bands according to human visual sensitivity. Luminance sub-bands receive enhanced processing for edge detection, while chrominance sub-bands use simplified processing, optimizing local contrast detection without uniformly complicating the entire processing pipeline.
2Manufacturing precision
If sub-band decomposition based on HVSC is implemented, then local image processing accuracy is improved, but computational complexity increases
Solution Approach 1:
By segmenting the image into sub-bands that align with human visual system characteristics, the decomposition enables targeted processing only where needed, reducing redundant computations while improving local accuracy.
Solution Approach 2:
Enhanced processing is applied selectively to luminance sub-bands where human visual sensitivity is highest, while chrominance sub-bands receive minimal processing. This partial action approach improves critical regions without uniformly increasing computational complexity across the entire image.
3Productivity
If conventional histogram equalization is used, then processing speed is maintained, but visual naturalness and edge detection quality deteriorate
Solution Approach 1:
The method segments the image into multiple sub-bands and processes them differently, allowing edge detection enhancement in luminance regions while maintaining processing efficiency through selective application of enhancement algorithms rather than global processing.
Solution Approach 2:
Edge detection and contrast enhancement are applied locally to luminance sub-bands where human visual sensitivity is highest, while chrominance sub-bands maintain their original characteristics. This local quality approach improves edge detection quality without requiring computationally intensive global processing.
Data Source
AI summary
Methods and apparatus for image processing include performing visual sub-band decomposition of an image using human visual system characteristics to generate a plurality of sub-band decomposed images, independently processing the plurality of sub-band decomposed images with at least one application, and fusing the independently processed sub-band decomposed images to reconstruct an output image.