Adaptive Intra-Prediction Filtering for Image Encoding Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image encoding/decoding methods are limited in efficiency due to the lack of varied filtering techniques, leading to increased costs in transmitting and storing high-resolution and high-quality image data.
Innovation Solution
The method involves determining the feature of the area where a reference sample is included, such as homogeneous, edge, or false edge areas, and applying appropriate filters like smoothing or edge-preserving filters, and excluding noise samples, to improve encoding/decoding efficiency by selecting the right filter based on the block size, form, prediction mode, and pixel component.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional filtering methods are used in image encoding/decoding, then device complexity is reduced, but encoding/decoding efficiency and image quality deteriorate
Solution Approach 1:
The patent applies parameter changes by dynamically adjusting filtering parameters (filter type, filter strength, filter direction) based on local image characteristics such as gradient magnitude, variance, and edge detection results. This allows the filtering process to adapt to different regions (smooth areas, edge areas, textured areas) thereby improving encoding efficiency without requiring a completely complex device architecture
Solution Approach 2:
The patent implements dynamics by making the filtering process adaptive and dynamic rather than static. The filtering parameters are determined dynamically based on local image statistics and characteristics, allowing the system to optimize encoding efficiency for each specific region while maintaining manageable device complexity through algorithmic adaptability
2Manufacturing precision
If high-resolution and high-quality image data is transmitted or stored using conventional methods, then image quality is improved, but transmission and storage costs increase
Solution Approach 1:
The patent applies local quality by differentiating filtering operations based on local image characteristics. Different filtering strategies are applied to different regions: smooth regions receive stronger filtering, edge regions receive edge-preserving filtering, and textured regions receive minimal filtering. This localized approach maintains high image quality where needed while reducing data complexity in appropriate regions, thereby lowering transmission and storage costs
Solution Approach 2:
The patent changes filtering parameters adaptively based on local image statistics to optimize the balance between image quality and compression efficiency. By adjusting filter strength, type, and direction according to local characteristics, the system maintains high quality where necessary while achieving better compression where possible, reducing overall transmission and storage requirements
3Manufacturing precision
If conventional filtering methods are used, then device complexity is reduced, but prediction efficiency and image quality deteriorate due to increased artifacts
Solution Approach 1:
The patent applies segmentation by dividing the image into different regions based on local characteristics (smooth regions, edge regions, textured regions) and applying appropriate filtering methods to each segment. This segmentation strategy improves prediction efficiency by ensuring each region receives the most suitable filtering treatment while keeping the overall system complexity manageable through modular processing
Solution Approach 2:
The patent changes filtering parameters based on detected image features such as gradient magnitude, variance, and edge presence. This adaptive parameter adjustment improves prediction efficiency by optimizing filtering for each local region while avoiding the need for a large variety of fixed filtering techniques, thereby managing device complexity effectively
Data Source
AI summary
The present invention relates to an image encoding and decoding method. An image decoding method for the same includes: determining a reference sample of a current block; performing filtering for the reference sample on the basis of a feature of an area where the reference sample is included; and performing intra-prediction by using the reference sample for which filtering is performed.