Super Resolution Interpolation Edge Smoothing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Digital zoom systems in digital cameras introduce significant image aberrations such as aliasing, blurring, and haloing, particularly at the edges of objects, which degrade the quality of the magnified image.
Innovation Solution
A single-frame based super resolution interpolation system that includes a super resolution engine with image region selection, resolution enhancement, and edge enhancement blocks to digitally zoom images without producing significant aberrations, using processes like bicubic interpolation and edge smoothing to improve edge quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If digital zoom systems process the image to increase resolution, then the field of view is narrowed and magnification is achieved, but significant image aberrations such as aliasing, blurring, and haloing are introduced
Solution Approach 1:
The patent applies edge smoothing processing to the input image before performing resolution enhancement. This preliminary action prepares the image data by reducing high-frequency noise and potential aliasing artifacts that would be amplified during subsequent upscaling operations, thereby preventing rather than correcting aberrations
Solution Approach 2:
The patent applies different processing strategies to different regions of the image. Edge regions undergo specialized edge-preserving smoothing while non-edge regions receive different treatment, allowing the system to maintain sharp object boundaries while still achieving overall resolution enhancement without uniform aberrations across the entire image
2Object-affected harmful factors
If optical zoom systems narrow the field of visible radiation, then magnification is achieved without significant image aberrations, but the field of view becomes narrower
Solution Approach 1:
The patent creates a digital copy of the optical zoom effect by using image processing algorithms to synthesize high-resolution details from a single lower-resolution frame. This digital copying approach replicates the magnification function of optical zoom while allowing the full sensor area to be utilized, thus maintaining a wider field of view
Solution Approach 2:
The patent replaces the mechanical optical zoom mechanism with a computational image processing system. Instead of physically moving lens elements to achieve magnification, the system uses algorithms to interpolate and enhance image details, eliminating the need for mechanical adjustment while achieving similar visual results
3Shape
If digital zoom systems magnify the captured image, then the field of view is narrowed with magnification effect, but aliasing and jagged edges are introduced
Solution Approach 1:
The patent changes the processing parameters dynamically based on local image characteristics. Edge detection algorithms identify regions requiring special treatment, and adaptive smoothing parameters are applied differently across the image - stronger smoothing in uniform regions and edge-preserving smoothing at boundaries, thereby maintaining sharp edges during magnification
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A digital camera system for super resolution image processing is provided. The digital camera system includes a resolution enhancement module configured to receive at least a portion of an image, to increase the resolution of the received image, and to output a resolution enhanced image and an edge extraction module configured to receive the resolution enhanced image, to extract at least one edge of the resolution enhanced image, and to output the extracted at least one edge of the resolution enhanced image, the at least one edge being a set of contiguous pixels where an abrupt change in pixel values occur. The digital camera system also includes an edge enhancement module configured to receive the resolution enhanced image and the extracted at least one edge, and to combine the extracted at least one edge or a derivation of the extracted at least one edge with the resolution enhanced image.