Multivalue Image Compression via Layer Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image compression methods for multivalue images, such as color images, often degrade image quality when trying to achieve high compression rates, especially when dealing with high contrast edges, and require accurate image segmentation which increases computation load and memory usage.
Innovation Solution
The method segments a multivalue image into multiple layer images, classifies pixel attributes, and applies suitable compression techniques to each layer, allowing for high compression rates without degrading image quality by adjusting pixel values and using different compression methods for distinct image sections.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If lossy compression method (e.g., JPEG) is applied to multivalue image, then compression rate is improved, but image quality deteriorates
Solution Approach 1:
The multivalue image is segmented into multiple layer images based on pixel value ranges. Different compression methods are applied to different layers: lossless compression for layers containing important features (foreground, edges) and lossy compression for background layers. This segmentation allows achieving high compression rates while preserving image quality in critical areas.
2Manufacturing precision
If different compression methods are applied to different sections of multivalue image, then image quality is improved, but computation load increases
Solution Approach 1:
The patent applies local quality principle by assigning different compression qualities to different spatial regions and layer types. Foreground layers with important visual information use lossless compression, while background layers use lossy compression. This localized approach improves overall image quality without requiring uniform high-quality processing across the entire image, thereby reducing total computation load.
3Productivity
If highly accurate image segmentation is performed to achieve high compression rate without degrading quality, then compression performance is improved, but memory space requirement increases
Solution Approach 1:
The patent transitions from spatial segmentation to spectral/value-based segmentation by dividing the image into multiple layer images based on pixel value ranges. This dimensional change allows processing different value ranges separately with appropriate compression methods, achieving high compression rates without requiring complex spatial segmentation algorithms that would consume excessive memory.
Data Source
AI summary
An apparatus, method, system, computer program and product, each capable of compressing a multivalue image with control of memory space requirement.