Threshold Value Matrix Generation for Blue Noise Dithering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The blue noise dither process requires extensive computation for large threshold value matrices and often results in image quality degradation due to specific patterns like chequered patterns, leading to increased processing time and reduced image quality.
Innovation Solution
A method involving a computation device that creates a threshold value matrix using a dither process by generating and modifying dot patterns based on an initial dot pattern, employing error diffusion methods to increase or decrease dot rates, and arranging gradation values as threshold values, thereby reducing computation time and improving image quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If a large threshold value matrix is used in the blue noise dither process, then the dot distribution uniformity is improved, but the processing time increases significantly
Solution Approach 1:
The patent divides the threshold value matrix into multiple smaller sub-matrices (e.g., 8x8, 16x16, or 32x32 blocks). Each sub-matrix is processed independently to create the final threshold value matrix. This segmentation allows the system to achieve uniform dot distribution across the entire image while significantly reducing the computational burden and processing time compared to creating a single large matrix.
2Manufacturing precision
If the blue noise dither process is used to improve dot distribution, then image quality is improved, but specific patterns like chequered patterns appear causing quality degradation
Solution Approach 1:
The patent applies different threshold value matrices to different regions of the image based on local characteristics. By analyzing the local content and adjusting the threshold value matrix accordingly, the system can maintain uniform dot distribution while avoiding the formation of unwanted patterns like chequered patterns in specific areas of the image.
3Productivity
If a large threshold value matrix is used to maintain high-speed processing, then processing speed is maintained, but computation time for matrix creation increases
Solution Approach 1:
The patent pre-calculates and stores multiple threshold value matrices of different sizes (e.g., 8x8, 16x16, 32x32) before the actual halftone processing begins. During processing, the system selects and applies the appropriate pre-computed matrix, eliminating the need to compute large threshold value matrices in real-time and significantly reducing matrix creation time while maintaining high processing speed.
Data Source
AI summary
In a first process, initial dot pattern of a predetermined dot rate “a” is created in a first process, and a gradation value corresponding to the dot rate “a” is arranged as a threshold value in the position of the pixel of the threshold value matrix corresponding to the initial dot pattern. Then in a second process, the initial dot pattern or the dot pattern having occurred prior to the second computation is used as the dot pattern of dot rate “b”, and new dots including the dots of the dot pattern of dot rate “b” is generated or any of the dots is removed from the dot pattern of the dot rate “b” by the error diffusion method. Thus, the dot pattern of the next dot rate “b′” with its dot rate having been increased or decreased in the aforementioned procedure is generated.