Psychovisual Modulation Image Processing Speed Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image processing methods for color images are inefficient due to increased processing time when applying psychovisual modulation (PVM) techniques, particularly when factorizing non-negative matrix factorization equations, which affects the speed of generating output images from basis element frames.
Innovation Solution
The method determines chrominance and luminance information of input color images, iteratively solving equations to obtain modulation weights and bases of basis element frames, focusing on luminance information first to improve processing speed, and then converts these into output images using non-negative matrix factorization equations with constraints.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If psychovisual modulation (PVM) techniques are applied to color images using conventional methods, then visual quality is maintained, but processing time increases significantly
Solution Approach 1:
The patent segments the color image processing into separate luminance and chrominance components. By decomposing the color image into Y (luminance) and UV (chrominance) channels and processing them independently through separate factorization operations, the method reduces computational complexity while maintaining visual quality, as the human visual system is more sensitive to luminance variations than chrominance variations.
Solution Approach 2:
The patent extracts and prioritizes luminance information from color images for factorization. By taking out the luminance component (Y channel) and performing PVM factorization primarily on this component, the method achieves faster processing speed since luminance contains the majority of visual information, while chrominance is processed with reduced complexity.
2Loss of information
If conventional factorization methods are used for color images, then complete color information is preserved, but processing speed decreases
Solution Approach 1:
The patent applies different processing quality levels to different color components. Luminance information (Y channel) is processed with high quality and detailed factorization, while chrominance information (UV channels) is processed with lower quality and reduced factorization complexity, matching the human visual system's differential sensitivity to these components.
Solution Approach 2:
The patent changes the processing parameters for different color channels by applying different factorization complexities. The luminance channel undergoes full PVM factorization with multiple basis elements, while chrominance channels use simplified factorization with fewer basis elements, optimizing the balance between information preservation and processing speed.
Data Source
AI summary
An image processing method includes: determining chrominance information and luminance information of an input color image; iteratively solving one or more equations to obtain bases of the luminance and modulation weights of a plurality of basis element frames of the input color image based on the luminance information; determining bases of the chrominance of the basis element frames based on the modulation weights and the chrominance information; and converting the bases of the luminance and the bases of the chrominance of the basis element frames into an output image.