Image Downscaler Tile Partitioning to Reduce Visual Artifacts
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Image scaling techniques result in visual artifacts, particularly blocking artifacts, due to the need for extrapolation of pixel values in upscaled images, which are less accurate than interpolation and can degrade image quality.
Innovation Solution
The method involves partitioning images into tiles based on a base tile size and scaling factors to ensure that all reconstructed pixel values in the upscaled image can be interpolated from two or more preserved pixel values, thereby reducing the need for extrapolation and minimizing visual artifacts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If image scaling is performed using conventional downscaling techniques, then bandwidth and storage requirements are reduced, but visual artifacts such as blocking artifacts appear in the upscaled image
Solution Approach 1:
The image is divided into multiple tiles, each of which is downscaled independently. This segmentation allows for selective preservation of pixel values at tile boundaries, ensuring continuity and reducing blocking artifacts when tiles are reconstructed and assembled during upscaling.
Solution Approach 2:
Pixel values at tile boundaries are preserved in advance during the downscaling process. By proactively retaining these boundary pixel values before upscaling, the system ensures that extrapolation can be avoided during reconstruction, thereby preventing blocking artifacts.
2Productivity
If pixel values are extrapolated during upscaling to reconstruct the original image, then the upscaled image can be generated from downscaled data, but the reconstructed pixel values are less accurate and produce visual artifacts
Solution Approach 1:
The system performs preliminary preservation of pixel values at tile boundaries during downscaling. This advance preparation ensures that when upscaling occurs, the necessary pixel values are already available, eliminating the need for extrapolation and thereby maintaining high pixel value accuracy.
Solution Approach 2:
Pixel values at tile boundaries are copied and preserved during the downscaling process. Instead of losing these critical values, the system creates copies that are retained and used during upscaling to accurately reconstruct the original image without requiring extrapolation.
3Device complexity
If the entire image is processed as a single unit during downscaling, then the process is simpler, but the proportion of extrapolated pixel values increases during upscaling
Solution Approach 1:
The image is segmented into multiple tiles for independent processing during downscaling. While this increases processing complexity compared to handling the entire image as one unit, it enables the preservation of boundary pixel values in each tile, which significantly reduces the need for extrapolation during upscaling and improves overall pixel value accuracy.
Data Source
AI summary
This disclosure provides methods, devices, and systems for image scaling. The present implementations more specifically relate to downscaling techniques that preserve a subset of pixel values in a digital image so that all reconstructed pixel values in the upscaled image can be interpolated based on two or more preserved pixel values. In some aspects, an image downscaler may partition a digital image into a number of image tiles based on a base tile size and one or more scaling factors associated with a downscaling operation used to downscale each image tile. The base tile size indicates the dimensions of the downscaled tiles. The one or more scaling factors indicate distances between the pixel values preserved as a result of one or more iterations, respectively, of the downscaling operation, where each iteration preserves a smaller subset of the pixel values from the original image (as a lower resolution tile).