Image Encoding Apparatus Residual Block Transformation Decision
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image encoding methods, such as those used in MPEG and HEVC standards, face inefficiencies in deciding whether to transform residual blocks into a frequency domain, leading to complex reconstruction processes and suboptimal compression performance.
Innovation Solution
An image encoding method based on a System on Chip (SoC) that calculates and compares values in both the space domain and transform domain to determine whether to transform residual blocks, using formulas or thresholds to decide on transformation, thereby optimizing encoding decisions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If complex reconstruction is fully performed to determine whether to skip transforming the residual block, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent performs a preliminary simplified evaluation using SAD (Sum of Absolute Differences) on the residual block before committing to full reconstruction. This preliminary action provides an early indication of whether transformation should be skipped, avoiding the need to always perform complex full reconstruction for decision-making purposes.
Solution Approach 2:
The decision process is segmented into multiple stages: first a simple SAD calculation is performed, then based on that result, a more sophisticated evaluation may be performed. This segmentation allows the system to achieve accurate decision-making while avoiding unnecessary computational complexity in cases where the simple evaluation is sufficient.
2Manufacturing precision
If transforming the residual block into frequency domain is always performed, then manufacturing precision is improved, but productivity decreases
Solution Approach 1:
The patent implements a dynamic transformation decision mechanism that adapts to the characteristics of each residual block. Based on SAD calculations and other metrics, the system dynamically determines whether transformation should be applied to each block, rather than uniformly applying transformation to all blocks. This dynamic approach optimizes the balance between encoding quality and processing speed.
Solution Approach 2:
Different transformation strategies are applied to different regions (blocks) based on their local characteristics. Blocks that benefit from transformation are transformed, while blocks that do not benefit have transformation skipped. This local quality approach ensures that transformation is applied only where it improves encoding quality, maximizing productivity without sacrificing necessary precision.
3Productivity
If transformation is skipped based on simple criteria, then productivity is improved, but measurement precision deteriorates
Solution Approach 1:
The system uses feedback from SAD calculations and other metrics to guide the transformation decision. By continuously evaluating the residual block characteristics and using this feedback to determine whether transformation should be applied, the system achieves accurate decision-making without requiring full reconstruction, thus maintaining both productivity and measurement precision.
Data Source
AI summary
An image encoding apparatus based on a system on chip (SoC) which encodes a residual block of a current block including a first value calculator configured to calculate a first value for the residual block in a space domain, a comparator configured to decide whether to transform the residual block into a transform domain according to a size of the first value, and a transform unit configured to transform the residual block into the transform domain according to a decision on whether to transform.