Nonlinear Image Resizing Filter for Sharp Edge Preservation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image resizing techniques using linear filters often result in blurry edges due to inappropriate mixing of pixel values from both sides of a sharp edge, leading to noise and ringing artifacts, which existing methods attempt to fix through sharpening or edge detection.
Innovation Solution
A method employing nonlinear filtering that weights the contributions of neighboring pixels based on their difference from the central pixel, using functions like Gaussian, to reduce or eliminate the mixing of pixel values and preserve sharp edges during image resizing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If linear kernel filters are used for image resizing, then the resizing operation is fast and simple, but sharp edges become blurred due to mixing pixels from both sides of the edge
Solution Approach 1:
The patent changes the filtering parameter from linear to nonlinear by introducing an edge-aware kernel that dynamically adjusts its behavior based on local image characteristics. The kernel modifies its weighting function to reduce mixing across edges while maintaining smooth blending in homogeneous regions, thus preserving edge sharpness without sacrificing resizing performance
Solution Approach 2:
The patent makes the kernel dynamic by allowing it to adapt its parameters based on local edge detection. The edge-aware kernel continuously adjusts its weighting scheme according to the detected edge strength and orientation in different regions, enabling it to preserve sharp edges where needed while maintaining efficient resizing elsewhere
2Manufacturing precision
If the kernel size is reduced to one pixel to avoid mixing pixels across edges, then edge mixing is eliminated, but the output becomes jagged due to nearest-neighbor resampling
Solution Approach 1:
The patent applies local quality by making the kernel behavior location-dependent. In regions near detected edges, the kernel uses a smaller effective support to avoid mixing pixels across the edge boundary. In homogeneous regions away from edges, the kernel uses its full support to maintain smooth resampling. This local adaptation allows the system to achieve both edge accuracy and output smoothness simultaneously
3Manufacturing precision
If sharpening operations are applied to resized images to improve edges, then edge sharpness is enhanced, but noise and ringing artifacts are introduced
Solution Approach 1:
The patent applies preliminary action by preserving edge sharpness during the resizing operation itself rather than attempting to fix blurred edges afterward. The edge-aware kernel prevents edge mixing at the source during resampling, eliminating the need for subsequent sharpening operations that would otherwise introduce noise and ringing artifacts
Data Source
AI summary
One embodiment of the present invention provides a system that uses nonlinear filtering while resizing an image to preserve sharp detail. The system starts with an original image, which is a digital image comprised of a plurality of pixels. Next, the system resizes the original image. This involves first producing an initial resized image by using neighboring pixel values in the original image (or a conventional resizing operation) to produce initial values for subpixel locations in the original image. The system then applies a nonlinear filter using pixel values in the original and initial resized images to produce a resized output image. When this nonlinear filter updates pixel values, it weights the contributions of neighboring pixels nonlinearly. This non-linear weighting of neighboring pixels can be used to significantly reduce (or eliminate) the mixing pixel values from both sides of a sharp edge caused by conventional linear filters, and can thereby alleviate the problem of blurring sharp edges during image resizing operations.