Adaptive Image Filtering for Lossless Compression Bandwidth
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
High-speed production printers face challenges in processing high pixel density images efficiently due to high performance data computation and data transportation requirements, leading to bandwidth bottlenecks and distortion in image compression.
Innovation Solution
An image filtering technique that segments raster images into regions and adjusts filtering based on an adjustable filter parameter, minimizing distortion in smaller regions while maximizing compression in larger regions, using a run-length encoding scheme to manage bandwidth limitations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If high pixel density images are processed at high speed, then printing performance is improved, but data computation and transportation challenges increase
Solution Approach 1:
The patent divides the raster image into multiple regions and processes each region independently with adaptive filtering. This segmentation allows parallel processing and reduces the computational burden on any single processing unit, enabling high-speed printing while managing data complexity effectively.
Solution Approach 2:
The patent applies different filter parameters to different regions of the image based on local characteristics. Each region is evaluated and assigned an appropriate filter strength, allowing high-speed processing of simple regions while applying more sophisticated processing only where needed, thus balancing speed and quality.
2Quantity of substance
If image compression is applied to reduce bandwidth, then data transportation is improved, but image distortion increases
Solution Approach 1:
The patent uses dynamic filter parameters that are adjusted based on the compression ratio achieved and the visual importance of each region. This dynamic adaptation allows the system to maximize compression while maintaining image quality in critical areas, resolving the contradiction between bandwidth reduction and distortion control.
Solution Approach 2:
The patent implements a feedback mechanism where the compression results are evaluated and used to adjust filtering parameters for subsequent processing. This closed-loop approach ensures that compression is optimized without excessive distortion, as the system learns from previous compression outcomes and adapts accordingly.
3Manufacturing precision
If adaptive filtering is applied to each region, then image quality is improved, but processing complexity increases
Solution Approach 1:
The patent changes the filter parameters adaptively based on region characteristics and compression requirements. By adjusting parameters such as filter strength and type according to local image properties, the system achieves high image quality without requiring uniformly complex processing across the entire image, thus managing overall processing complexity.
Data Source
AI summary
Systems, methods, software for image filtering. In one embodiment, a system receives a raster image comprising an array of pixels, divides the raster image into regions of pixels, and identifies a region size limit for the regions. The system performs image filtering on each region with an image filter that operates based on a filter parameter that is adjustable between a first value where minimal filtering is performed and a second value where maximum filtering is performed. For the image filtering, the system measures a compressed region size of the region when compressed with a run-length encoding scheme, computes the filter parameter for the region based on the compressed region size and the region size limit, and applies the image filter on the region based on the filter parameter computed for the region to generate a filtered region having increased redundant patterns of pixel values.