Adaptive Image Filtering Using Bit-Shift Multiplier Coefficients
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The implementation of adaptive filtering with multiplicative filter coefficients is costly due to the large number of multiplication operations required, especially in dedicated hardware, which complicates efficient processing.
Innovation Solution
The solution involves restricting the values of adaptive multiplier coefficients to a set of allowed values with a predetermined number of binary digits, ensuring at least one 'zero' in their binary representation, thereby simplifying multiplication operations and enhancing processing efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If adaptive filtering with multiplicative filter coefficients is implemented, then filtering performance is improved, but computational complexity and hardware cost increase due to large number of multiplication operations
Solution Approach 1:
The patent changes the parameter representation of filter coefficients from general floating-point numbers to a restricted set of values with specific binary characteristics (at most one '1' bit). This parameter transformation converts complex multiplication operations into simpler bit-shift operations, reducing computational complexity while preserving filtering performance
Solution Approach 2:
The patent replaces the mechanical multiplication operation with a bit-shift operation. By restricting coefficients to have at most one '1' bit in their binary representation, the multiplication coefficient becomes a power of two, which can be implemented as a simple bit-shift rather than a complex multiplier, significantly reducing hardware complexity
2Reliability
If adaptive filtering with multiplicative filter coefficients is implemented, then filtering performance is improved, but hardware implementation cost increases
Solution Approach 1:
The patent replaces expensive multiplication hardware with simpler bit-shift hardware. By transforming coefficients into a restricted set where at most one bit is set, the multiplication operation is substituted with a bit-shift operation that requires no multiplier hardware, significantly reducing hardware implementation cost
Solution Approach 2:
The patent uses a simplified representation of coefficients that can be easily generated and applied. Instead of using complex floating-point coefficients that require expensive multiplication hardware, the system uses simplified binary coefficients (powers of two) that can be implemented with simple bit-shift operations, effectively using 'cheap' computational operations instead of expensive hardware
Data Source
AI summary
An apparatus and a method are provided for filtering reconstructed images, in particular, video images, with adaptive multiplicative filters. The efficiency of the filtering operation is increased by restricting the allowable values of the filter coefficients to those that have only a limited number of “ones” in the binary representation.