Adaptive De-Block Filtering for Block-Based Image Decoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Decoding systems, such as those for MPEG 2 and H.264/AVC, face challenges in suppressing block distortion during image decoding due to the absence of block boundary strength data and quantization parameter information in encoded data, which are essential for de-block filtering.
Innovation Solution
An image processing system that includes a controlling unit to select appropriate filtering content based on encoding types and a filtering unit to apply filtering to block image data, utilizing reversible decoding, inverse quantization, and inverse orthogonal transformation to generate reconfigured image data for effective de-block filtering.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If de-block filtering is applied using block boundary strength data Bs and quantization parameter QP, then block distortion is suppressed and image quality is improved, but encoded data size increases due to adding Bs and QP information
Solution Approach 1:
The decoding system generates the filtering parameters (block boundary strength data and quantization parameter) by itself from the encoded data without requiring these parameters to be explicitly transmitted. The system uses the encoding type information already present in the encoded data to determine appropriate filtering content, making the system self-sufficient and avoiding additional data transmission.
Solution Approach 2:
The encoding type information acts as an intermediary that bridges the gap between the encoded data and the filtering process. Instead of directly transmitting filtering parameters, the system uses encoding type as a mediator to indirectly determine the appropriate filtering content, thus achieving filtering without increasing encoded data size.
2Quantity of substance
If de-block filtering is not applied due to missing filtering information, then encoded data size remains small, but block distortion remains and image quality deteriorates
Solution Approach 1:
The system performs self-service by generating necessary filtering parameters from existing encoded data through inverse quantization and inverse orthogonal transform processes, eliminating the need for external parameter transmission while maintaining filtering capability.
Solution Approach 2:
The system performs preliminary inverse quantization and inverse orthogonal transform operations on the encoded data to generate the filtering parameters before applying de-block filtering. This preliminary processing enables the filtering to be performed with adequate parameters without requiring additional data transmission.
3Manufacturing precision
If adaptive filtering based on encoding types is implemented, then block distortion is suppressed without increasing encoded data size, but system complexity increases due to multiple processing circuits
Solution Approach 1:
The decoding system is designed with multi-functionality, where the same circuit performs multiple functions: reversible decoding, inverse quantization, inverse orthogonal transform, and de-block filtering. This universal approach allows the system to generate filtering parameters from the encoded data itself, achieving distortion suppression without proportionally increasing system complexity.
Solution Approach 2:
The patent merges multiple processing functions into an integrated decoding pipeline. The inverse quantization circuit and inverse orthogonal transform circuit work in conjunction with the de-block filtering circuit, combining parameter generation and filtering operations into a unified system that achieves effective distortion suppression without excessive complexity increase.
Data Source
AI summary
There is provided an image processing system and an image processing method able to suppress block distortion in the case of decoding image data encoded in unit of blocks. A controlling unit selects a filtering content to be applied to the block image data based on the encoding types of the block image data to be filtered, and a filtering unit applies filtering to the block image data to be processed according to the filtering content selected by the controlling unit.