Block Image Compression Using Multi-Level Difference Tables
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing lossy compression techniques for image data in low-power and low-cost computing systems, such as mobile devices, either fail to achieve a guaranteed compression ratio or introduce unacceptable quality degradation, making them unsuitable for use in small, low-cost, and low-power systems.
Innovation Solution
A lossy compression method using a multi-level difference table to compress image data blocks, ensuring a fixed compression ratio (e.g., 2:1) without significant quality loss, implemented through simple integer operations in dedicated hardware, such as GPUs or CPUs, without requiring complex operations or internal buffering.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If current lossy compression techniques are used to achieve guaranteed compression ratio, then compression ratio is improved, but image quality deteriorates significantly and device complexity increases
Solution Approach 1:
The image data is divided into 4x4 pixel blocks for independent compression processing. Each block is handled separately through the multi-level difference table method, allowing localized optimization of compression quality while maintaining overall image fidelity.
Solution Approach 2:
The patent transforms the compression approach by introducing a multi-level difference table structure that operates on pixel value differences rather than direct pixel values. This dimensional transformation enables more efficient entropy encoding and achieves guaranteed compression ratios with minimal quality loss.
2Quantity of substance
If current lossy compression techniques are used to achieve guaranteed compression ratio, then compression ratio is improved, but device complexity increases due to complex operations and internal buffering
Solution Approach 1:
The patent extracts and removes the need for complex internal buffering operations by using a streamlined multi-level difference table approach. The compression is achieved through simple arithmetic operations on pixel differences followed by entropy encoding, eliminating the need for complex intermediate storage and manipulation.
Solution Approach 2:
The patent employs temporary difference tables that are computed and discarded for each 4x4 block, rather than maintaining complex persistent buffer structures. This disposable approach to intermediate data structures reduces device complexity while achieving the same compression goals.
3Loss of information
If data compression is not applied, then image quality is maintained, but memory bandwidth and power consumption increase
Solution Approach 1:
The patent changes the representation parameters of image data by encoding pixel differences rather than absolute pixel values. This parameter transformation reduces the entropy of the data, enabling more efficient compression while maintaining visual quality, thereby reducing memory bandwidth requirements and associated power consumption.
4Loss of information
If data compression is not applied, then image quality is maintained, but memory bandwidth increases
Solution Approach 1:
The patent performs preliminary differential encoding on image blocks before transmission or storage. By pre-processing the data to encode differences rather than absolute values, the compression is achieved upfront, reducing the bandwidth required for subsequent memory operations without compromising image quality.
Data Source
Figure 1~2
Figure 3a~3b
Figure 4
AI summary
There is provided a computer-implemented method and a compression unit for performing lossy compression on a block of image data in accordance with a multi-level difference table. The block of image data comprises a plurality of image element values, wherein each level of the multi-level difference table comprises a plurality of entries. An origin value for the block of image data is determined. A level within the multi-level difference table for the block of image data is determined. For each image element value in the block of image data, one of the entries at the determined level within the multi-level difference table is selected. A compressed block of data for the block of image data is formed, wherein the compressed block of data comprises: (i) data representing the determined origin value, (ii) an indication of the determined level, and (iii) for each image element value in the block of image data, an indication of the selected entry for that image element value.