Image Processing Apparatus Reducing Texture Noise via Localized Pixel Concentration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for converting low resolution multi-level image data to high resolution image data often result in texture noise due to the use of dither threshold matrices that concentrate black pixels throughout the entire image, leading to reduced dot reproducibility and image quality.
Innovation Solution
An image processing apparatus that converts tone levels based on a threshold value, diffuses errors between tone levels using a diffusion coefficient pattern, and selectively concentrates black pixels by referring to output sequence categories and patterns stored in memory, allowing for enhanced dot reproducibility and image quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a dither threshold matrix is used to concentrate black pixels throughout the whole image data, then dot reproducibility is enhanced, but texture noise easily occurs
Solution Approach 1:
The patent applies local quality by differentiating processing based on pixel tone levels. Specifically, pixel concentration processing is selectively applied only to pixels with tone levels of 1 or 2 (darker pixels), while pixels with higher tone levels undergo different processing. This localized approach concentrates black pixels where needed for dot reproducibility while avoiding excessive concentration in lighter areas, thereby preventing texture noise generation.
Solution Approach 2:
The patent employs dynamics by introducing randomness in the output sequence of pixels during the conversion process. The output sequence category is determined dynamically based on surrounding pixel characteristics, and the actual output sequence pattern is selected with a predetermined probability. This dynamic, probabilistic approach prevents fixed pattern formation that would otherwise generate texture noise, while still maintaining the overall pixel concentration effect.
2Reliability
If black pixels are concentrated throughout the whole image data, then isolated black pixels are prevented from not reproducing in printing, but a certain pattern is produced
Solution Approach 1:
The patent introduces dynamics through probabilistic selection of output sequence patterns. Instead of deterministically assigning a fixed pattern to each pixel based on its tone level, the system selects from multiple possible patterns with predetermined probabilities. This randomness breaks the regularity that would otherwise create visible patterns, while the overall statistical distribution maintains the pixel concentration effect needed for dot reproducibility.
Solution Approach 2:
The patent applies local quality by determining output sequence categories based on local surrounding pixel characteristics. Each pixel's processing is adapted to its local context, preventing uniform pattern repetition across the entire image. This localized adaptation ensures dot reproducibility in each region while avoiding global pattern formation.
3Measurement precision
If pixel concentration processing is performed using a dither threshold matrix, then granularity and resolution are achieved, but the processing complexity increases
Solution Approach 1:
The patent applies segmentation by dividing the complex pixel concentration process into distinct stages: tone level conversion, output sequence category determination, and output sequence pattern selection. Each stage handles a specific aspect of the conversion, making the overall process more manageable and implementable. The segmentation also allows for optimized processing at each stage, reducing overall computational complexity while maintaining high resolution output.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The apparatus effectively enhances dot reproducibility and image quality by concentrating black pixels in low concentrated areas, reducing texture noise and improving the overall quality of converted image data.
Implementation Method 1
an error diffusing section to diffuse an error between the tone levels before and after the tone conversion in the tone converting section to a surrounding pixel of the target pixel, based on a diffusion coefficient pattern
Data Source
AI summary
Disclosed is an image processing apparatus including a tone converting section to convert a tone level of a target pixel in multi-level image data based on a threshold value of the tone level so that the number of tone levels is reduced; a resolution converting section to output a pixel block according to the tone level, to generate image data with higher resolution; and an error diffusing section to diffuse an error; and wherein when the converted tone level is a predetermined value or lower, the resolution converting section refers to an output sequence category of a black pixel in a pixel block of a surrounding pixel, selects an output sequence pattern belonging to an output sequence category which allows a black pixel in a pixel block of the target pixel and the surrounding pixel to be concentrated, and outputs a pixel block corresponding to the selected pattern.