Image Interpolation Using Multi-Stage Spatial Domain Filters
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image encoding and decoding methods face challenges in precisely predicting pixel values between integer pixel units, particularly in interpolating fractional pixel values for accurate image representation and compression.
Innovation Solution
The method involves selecting and using spatial-domain filters with shifted phases to transform and inverse transform pixel values, allowing for interpolation between integer and fractional pixel units, enhancing precision in image interpolation and compression.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional interpolation methods are used to generate fractional pixel values, then the process is simple, but the prediction precision is insufficient
Solution Approach 1:
The patent segments the interpolation process into multiple stages: first generating fractional pixel values using a first filter, then generating intermediate pixel values using a second filter, and finally generating target pixel values using a third filter. This multi-stage segmentation enables higher prediction precision by progressively refining the interpolation results through different filtering operations.
Solution Approach 2:
The patent introduces an additional processing dimension by applying multiple filters (first, second, and third filters) with different characteristics to the pixel values at different stages. This dimensional expansion of the filtering process allows the system to capture more nuanced spatial relationships and achieve superior prediction precision without oversimplifying the interpolation.
2Measurement precision
If multiple filters are applied in sequence to improve interpolation accuracy, then the prediction precision improves, but the processing time increases
Solution Approach 1:
The patent applies different filters (first, second, and third filters) with distinct characteristics to different stages of the interpolation process. Each filter is optimized for specific local requirements: the first filter handles initial fractional pixel generation, the second filter refines intermediate values, and the third filter produces final accurate pixel values. This localized optimization of filter quality at each stage achieves high accuracy while managing processing time through targeted computations.
Data Source
AI summary
Provided are a method and apparatus for interpolating an image. The method includes: selecting a first filter, from among a plurality of different filters, for interpolating between pixel values of integer pixel units, according to an interpolation location; and generating at least one pixel value of at least one fractional pixel unit by interpolating between the pixel values of the integer pixel units by using the selected first filter.