Edge-Adaptive Recursive De-Ringing Filter for Image Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional linear low-pass filters used in image and video processing distort visual content and fail to effectively reduce multi-dimensional ringing effects, such as 'mosquito noise', especially in compressed formats, and are often specific to certain compression standards, making them impractical for handling various video and imaging standards.
Innovation Solution
An edge-adaptive and recursive non-linear filtering method that detects edges and applies a filtering process based on an edge influence function, adjusting pixel values adaptively for horizontal, vertical, and diagonal angles to reduce ringing effects independently of compression standards.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If conventional linear low-pass filters are used to reduce ringing effects, then the ringing artifacts are filtered out, but the visual content is distorted and image quality deteriorates
Solution Approach 1:
The patent applies different filtering operations to different regions of the image based on local characteristics. Edge pixels identify regions with high-frequency content where ringing occurs, and only these specific regions undergo filtering while preserving the rest of the image quality. This local approach prevents global distortion of visual content.
Solution Approach 2:
The filtering process is adaptive and dynamic rather than static. The filter dynamically adjusts its behavior based on whether a pixel is identified as an edge pixel or not, and recursively processes pixels based on their influence from neighboring edge pixels. This dynamic adaptation allows the filter to reduce ringing only where necessary without distorting other image regions.
2Object-affected harmful factors
If conventional linear low-pass filters are used to filter ringing effects, then the filtering is applied uniformly, but high-frequency content transitions are degraded and resolution is reduced
Solution Approach 1:
The patent identifies edge pixels that correspond to high-frequency transitions and applies filtering only to these specific locations and their influenced neighbors. Non-edge regions with important high-frequency content remain untouched, preserving both the transitions and the overall image resolution.
Solution Approach 2:
Instead of applying uniform filtering across the entire image, the patent applies partial filtering only to edge pixels and pixels influenced by them. This partial action approach filters out ringing artifacts in specific problematic regions while leaving the rest of the high-frequency content intact.
3Reliability
If conventional de-ringing filters are designed in a loop with compression standards, then they work for specific standards, but they cannot handle various image and video compression standards
Solution Approach 1:
The patent presents a universal de-ringing filter that operates independently of any specific compression standard. By detecting edge pixels based on local intensity variations and applying adaptive filtering, the method works effectively across different compression standards (JPEG, MPEG, etc.) without requiring standard-specific customization, achieving both reliability and versatility.
Solution Approach 2:
Instead of designing filters specific to each compression standard, the patent inverts the approach by creating a standard-agnostic filter that detects ringing artifacts based on their visual characteristics rather than their origin. This inverted methodology allows the same filter to handle multiple compression standards effectively.
4Device complexity
If conventional linear de-ringing filters are used, then the filtering process is simple, but multi-dimensional ringing effects and mosquito noise are difficult to reduce
Solution Approach 1:
The patent extends the filtering approach to multiple dimensions by considering horizontal, vertical, and diagonal edge relationships. The recursive filtering process propagates edge influences across different spatial dimensions, effectively addressing multi-dimensional ringing effects and mosquito noise that single-dimensional filters cannot handle.
Solution Approach 2:
The patent employs a recursive filtering process where the output of one filtering operation feeds back into subsequent operations. Edge pixels are identified, filtered, and their influence propagates to neighboring pixels, which may in turn become edge pixels themselves. This feedback mechanism allows the filter to progressively reduce multi-dimensional ringing effects through iterative refinement.
Data Source
AI summary
One or more systems and methods for edge-adaptive and recursive non-linear filtering of ringing effect on image or video data are disclosed in accordance with various embodiments of the invention. In one embodiment of the invention, a method for edge-adaptive and recursive non-linear filtering of ringing effect first involves detecting all edges within a process area of an image, including a direction of an edge slope and a value of an edge signal level. Then, if the process area is determined to be a “non-busy” area based on a “busyness” measure relative to empirically-defined threshold values, then an edge influence function subsequently determines whether to apply a de-ringing filter to a current pixel within the process area or not. Preferably, the de-ringing filter is edge-adaptive, non-linear, and recursive. The de-ringing filter can adjust the current pixel multiple times based on filter angles, adjacent pixels, and pixel transition levels.