Image Resampling Filters for Transparent Video Format Conversion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing technologies face challenges in achieving transparent up- and down-conversion between different sampling structures, particularly in multi-standard video processing systems, with prior filters exhibiting poor stop-band attenuation and aliasing issues, especially when converting between common horizontal and vertical sampling structures like 720, 960, and 1080 samples per line.
Innovation Solution
The development of optimized up- and down-conversion filters with identical frequency responses and reduced reconstruction errors, where every filtered sample is formed from a weighted sum of at least two input samples, and the filters are designed to minimize aliasing by optimizing the frequency response to have a null response at the lower sampling frequency, ensuring the combination of up-sampling and down-sampling filters is substantially transparent.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If conventional up- and down-conversion filters are used, then conversion between sampling structures can be achieved, but reconstruction errors and aliasing are significant
Solution Approach 1:
The patent applies parameter changes by optimizing the filter coefficients and frequency response characteristics. Specifically, the down-conversion filter is designed with a frequency response that has a null at the lower sampling frequency, and the coefficients are optimized to minimize reconstruction errors. This transforms the conventional filter design parameters to achieve superior reconstruction accuracy and aliasing rejection.
2Reliability
If asymmetric down-conversion filters are used to achieve reversibility, then mathematical transparency can be achieved, but filter complexity and implementation difficulty increase
Solution Approach 1:
The patent deliberately introduces asymmetry in the down-conversion filter design. The frequency response is made asymmetric by placing a null at the lower sampling frequency rather than at DC, and the coefficients are optimized asymmetrically to achieve reversibility. This controlled asymmetry enables mathematical transparency while maintaining implementation feasibility.
3Object-generated harmful factors
If filter aperture is increased to improve stop-band attenuation, then aliasing is reduced, but computational complexity and processing time increase
Solution Approach 1:
The patent optimizes the filter coefficients to achieve high stop-band attenuation with a moderate aperture size. By carefully selecting the coefficient values and optimizing the frequency response, the patent achieves superior aliasing rejection without requiring an excessively large filter aperture, thus maintaining reasonable processing speed.
Data Source
Figure 1
Figure 2~4
Figure 3
AI summary
First image data at a lower sampling frequency is up-sampled in a sampling ratio N:M to a higher sampling frequency in an up-sampling filter; and, second image data at the said higher sampling frequency is down-sampled in a sampling ratio M:N to the said lower sampling frequency in a down-sampling filter where the combination of the up- sampling filter and the down-sampling filter is substantially transparent andevery filtered sample is formed from a weighted sum of at least two input samples.