Adaptive Image Quantizer Matching for Compression and Quality
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image compression methods face challenges in achieving favorable context-dependent quantizer selection and efficient bandwidth reduction, often resulting in image artefacts or suboptimal quality loss.
Innovation Solution
A method for determining a quantizer with a specified number of levels by considering another quantizer with fewer levels, allowing for non-linear graduation and balancing quantizer values to control image data compression, and implementing multiple quantizers with varying levels to adapt to different image contexts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a quantizer with fewer levels is used to achieve higher compression, then bandwidth reduction is improved, but image quality deteriorates and artefacts occur
Solution Approach 1:
The patent applies dynamics by making the quantizer adaptive rather than static. The quantizer automatically adjusts its characteristics based on local image properties such as gradient magnitude and variance. This allows the system to use fewer quantization levels in smooth areas (achieving compression) while maintaining more levels in detailed areas (preserving quality), thus resolving the contradiction between bandwidth reduction and image quality.
Solution Approach 2:
The patent implements local quality by applying different quantization strategies to different regions of the image. Based on local statistical properties (gradient, variance), the quantizer adapts its level distribution locally. This means that important image features receive finer quantization while less important areas use coarser quantization, achieving both compression and quality preservation simultaneously.
2Manufacturing precision
If a quantizer with more levels is used to maintain image quality, then image quality is improved, but compression ratio deteriorates
Solution Approach 1:
The patent applies parameter changes by dynamically adjusting the quantization parameters (number of levels, level distribution) based on local image statistics. Instead of using a fixed high-level quantizer throughout the image, the system changes parameters locally according to gradient magnitude and variance. This allows the compression ratio to be optimized by using fewer levels where appropriate while maintaining quality where needed.
Solution Approach 2:
The quantizer is made dynamic by continuously adapting its parameters to local image content. The system calculates local statistics and adjusts quantization levels accordingly, transforming a static high-quality quantizer into a dynamic adaptive quantizer that achieves both quality and compression goals.
3Manufacturing precision
If context-dependent quantizer selection is implemented, then image quality is improved, but computational complexity increases
Solution Approach 1:
The patent applies preliminary action by pre-calculating or pre-determining quantizer characteristics based on local image statistics before actual quantization. The system computes gradient and variance measures in advance, uses these to select or configure appropriate quantizers, and then applies them. This preliminary analysis enables context-dependent selection without excessive computational overhead during the main processing stage.
Solution Approach 2:
The patent implements segmentation by dividing the image into local regions and applying different quantization strategies to each segment. By processing the image in manageable local blocks and determining quantizer parameters for each segment independently, the computational complexity is distributed and managed efficiently while achieving context-dependent optimization.
Data Source
AI summary
A method for determining a second quantizer for quantizing digital images, wherein the second quantizer is determined for a specified number of levels, which is at least two. For the determination, a first quantizer with a lower number of levels than the specified one is taken into consideration. Furthermore, a method for coding an image comprising a plurality of pixels, a computer-readable medium, an apparatus, which implements at least two quantizers as a digital circuit and a digital camera with such an apparatus is disclosed.