Image Processing Method for Ringing Reduction in Image Enlargement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing methods fail to effectively address ringing issues and maintain the Markov property during high-resolution image enlargement from low-resolution images, leading to distortion and quality deterioration, especially in non-high-definition content conversion.
Innovation Solution
The method calculates pixel difference sums and deviations between a pixel of interest and its surrounding pixels, adjusts these values using a constant, and applies a minimum coefficient to reduce ringing by ensuring generated pixel values fall within the neighborhood's value range, thereby securing the Markov property and enhancing image quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional interpolation techniques are used to increase pixel count, then the processing speed is high, but the image quality deteriorates with ringing artifacts and pseudo contours
Solution Approach 1:
The invention changes the fundamental parameter of pixel value generation from simple interpolation to solving optimization equations. By calculating pixel values that satisfy specific mathematical conditions (Equations 1-3), the method achieves both high processing speed and high image quality, eliminating ringing artifacts while maintaining Markov property.
2Measurement precision
If high-resolution techniques are applied to generate pixel values, then the resolution is increased, but ringing artifacts and distortion occur
Solution Approach 1:
The invention converts the harmful ringing artifacts into beneficial smooth transitions by carefully controlling pixel value ranges. By ensuring generated pixel values fall within the range of neighboring pixel values (Equation 3), the method eliminates ringing while maintaining the resolution enhancement benefit.
Solution Approach 2:
The method incorporates feedback mechanisms by using calculated pixel difference sums and deviations to guide the generation of new pixel values. The optimization equations use feedback from neighboring pixel values to adjust and constrain the generated values, preventing ringing artifacts.
3Device complexity
If pixel values are generated without constraining to neighborhood range, then computation is simpler, but the Markov property is not secured
Solution Approach 1:
The invention performs preliminary actions by pre-calculating pixel difference sums and deviations before generating final pixel values. This preliminary computation establishes the foundation for efficiently solving the optimization equations, ensuring both computational feasibility and Markov property satisfaction.
Data Source
AI summary
An image processing method is provided. Calculating sums of differences identified between a pixel value of a pixel of interest in an input image and pixel values of three pixels surrounding the pixel of interest. Calculating an average value of four pixel difference sums calculated by the pixel difference sum calculators. Calculating deviations between the average value and the four pixel difference sums calculated by the pixel difference sum calculators. Deriving a minimum coefficient from the candidates of coefficient calculated by a candidate coefficient calculator by using adjusted deviations derived from multiplying the deviations by a constant. Subtracting values derived from multiplying the adjusted deviations by the minimum coefficient from the pixel value of the pixel of interest in the input image and outputting values of four pixels in an enlarged image twice an original size of the input image in horizontal and vertical directions.