Image Processing Apparatus Dot Dispersibility N-gradation Multi-valued Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing methods, particularly in inkjet recording apparatuses, face challenges in maintaining image quality by preventing an increase in granularity, especially when dealing with small pixel value amplitudes in image regions.
Innovation Solution
An image processing apparatus is designed with a determination unit that assesses the difference between maximum and minimum pixel values in a unit region of interest and a generation unit that generates M-gradation quantization data based on N-gradation multi-valued data. If the difference is below a threshold, the generation unit ensures each pixel in the region has a value that is one of two consecutive M-gradation quantization values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If multi-value quantization is performed using a dither method with a threshold value matrix to improve dot dispersibility, then processing load is reduced compared to error diffusion method, but granularity increases in regions with small pixel value amplitudes
Solution Approach 1:
The patent applies different quantization strategies to different regions based on their characteristics. For regions with small pixel value amplitudes (low variability), a specialized handling approach is used that limits quantization levels to two consecutive values, preventing granularity. For other regions, standard dither quantization is applied. This local adaptation resolves the contradiction by maintaining high processing efficiency while improving image quality in problematic regions.
Solution Approach 2:
The patent dynamically changes the quantization parameter (number of output levels) based on the input region's pixel value amplitude characteristics. When the amplitude is below a threshold, the quantization is restricted to two consecutive levels instead of the full M-gradation range. This parameter adaptation prevents granularity in low-variability regions while maintaining the benefits of dither quantization elsewhere.
2Ease of operation
If standard M-gradation quantization is applied to all regions, then processing is simple and fast, but regions with small pixel value amplitudes exhibit increased granularity
Solution Approach 1:
The patent introduces a region-specific processing approach where regions with small pixel value amplitudes are identified and handled differently from other regions. This local quality differentiation maintains overall processing simplicity while applying specialized granulariry control only where needed, resolving the contradiction between ease of operation and granularity control.
Solution Approach 2:
The patent performs preliminary analysis of pixel value amplitude in each region before applying quantization. By pre-identifying regions with small amplitudes, the system can prepare appropriate quantization parameters in advance, avoiding granularity issues without complicating the overall processing flow. This preliminary action maintains ease of operation while ensuring quality control.
Data Source
AI summary
A determination unit determines whether a difference between a maximum value and a minimum value of pixel values of a plurality of pixels in a region of interest in N-gradation multi-valued data, where N≥4, representing the pixel values of the plurality of pixels is less than a determination threshold value. A generation unit generates M-gradation quantization data represented by an M-gradation quantization value, where 3≤M<N, based on the N-gradation multi-valued data and the determination. In a case where the determination unit determines that the difference is less than the determination threshold value, the generation unit generates the M-gradation quantization data so that each of the plurality of pixels corresponding to the region has a value that is one of two consecutive values of the M-gradation quantization values in the generated M-gradation quantization data.