Image Processing Apparatus High-Resolution Threshold Matrix
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image processing apparatuses require large-scale circuitry and increased computational costs due to the need for screen arithmetic circuits corresponding to the resolution conversion from input images to high-resolution threshold matrices, making practical use difficult.
Innovation Solution
An image processing apparatus that uses a high-resolution threshold matrix, where a weighted average of thresholds is calculated for each pixel of the input image, reducing the need for extensive circuitry by converting the high-resolution matrix to a low-resolution matrix matching the input image resolution, and randomly selecting weighting coefficients to suppress noise and enhance tone.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the input image is converted to the same resolution as the high-resolution threshold matrix, then screen processing quality is improved, but the number of screen arithmetic circuits required increases significantly
Solution Approach 1:
The patent divides the high-resolution threshold matrix into multiple blocks and processes only the relevant block corresponding to each input pixel, rather than converting the entire image to high resolution. This segmentation approach maintains processing quality while reducing circuit requirements.
Solution Approach 2:
The patent changes the approach from spatial resolution conversion (converting entire image to high resolution) to threshold value interpolation in the frequency domain. By calculating new threshold values through weighted averages of surrounding threshold values, the system achieves high-quality screen processing without requiring high-resolution conversion circuits.
2Measurement precision
If the resolution is converted from low to high resolution, then screen processing accuracy is improved, but computational cost increases
Solution Approach 1:
The patent pre-calculates and stores threshold values in a high-resolution threshold matrix during system initialization. During actual screen processing, the system retrieves and interpolates from this pre-computed matrix rather than performing full high-resolution conversion, significantly reducing real-time computational cost.
Solution Approach 2:
The patent creates a down-sampled version of the high-resolution threshold matrix that matches the input image resolution. This copied, lower-resolution threshold matrix is used for processing, maintaining accuracy while reducing the computational burden of handling full high-resolution data.
3Measurement precision
If a high-resolution threshold matrix is used, then image quality is improved, but the circuit scale becomes large
Solution Approach 1:
The patent segments the threshold matrix processing into block-level operations rather than full-image operations. By processing one block at a time and using local threshold interpolation, the circuit only needs to handle small portions of the image at high resolution, dramatically reducing the required circuit scale.
Solution Approach 2:
The patent changes the threshold values dynamically through weighted average calculations based on the input pixel position and surrounding threshold values. This parameter transformation approach allows the system to use a compact circuit that computes threshold values on-the-fly rather than storing and processing large high-resolution threshold matrices.
Data Source
AI summary
Disclosed is an image processing apparatus which uses a threshold matrix of a resolution higher than a resolution of an input image, the image processing apparatus including: a screen processing section which performs screen processing on an input image which is input; a matrix storage section which stores a threshold matrix of a resolution higher than a resolution of the input image; a threshold obtaining section which obtains a new threshold corresponding to each pixel of the input image based on a threshold of each cell composing the threshold matrix, wherein the screen processing section compares the pixel value of each pixel of the input image with the new threshold corresponding to each pixel of the input image obtained by the threshold obtaining section and generates a multivalue output image.