Adaptive Wavelet Encoding for 2D Signal Compression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing wavelet transform encoding methods struggle to determine the optimal encoding method for two-dimensional signals, as they either require collective or individual encoding of wavelet transform coefficients, leading to suboptimal compression ratios depending on the signal type.
Innovation Solution
A wavelet transform encoding apparatus that includes a coefficient group extracting unit for grouping wavelet transform coefficients and a mode determination reference coefficient extracting unit to compare the sum of absolute values with thresholds, determining whether to use multidimensional or individual coefficient encoding based on these values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If collective encoding method is used for wavelet transform coefficients, then encoding efficiency is improved for signals with similar coefficient patterns, but compression ratio deteriorates for signals with diverse coefficient distributions
Solution Approach 1:
The patent applies dynamics by making the encoding method adaptive rather than fixed. The encoding apparatus dynamically switches between collective encoding and individual encoding based on the actual characteristics of the wavelet transform coefficients. Specifically, it calculates the standard deviation of coefficient magnitudes in a block and selects the encoding method accordingly, allowing the system to adapt to different signal types and achieve optimal compression ratios.
Solution Approach 2:
The patent changes the encoding parameter (encoding method type) based on the statistical properties of the input coefficients. By calculating the standard deviation as a parameter and comparing it against thresholds, the system selects between different encoding strategies. This parameter-based adaptation resolves the contradiction by optimizing for compression ratio while maintaining encoding efficiency where applicable.
2Loss of information
If individual encoding method is used for wavelet transform coefficients, then compression ratio is improved for signals with diverse coefficient distributions, but encoding efficiency deteriorates compared to collective encoding
Solution Approach 1:
The system dynamically adjusts the encoding approach by evaluating coefficient characteristics before encoding. When individual encoding is determined to be more suitable (based on high standard deviation indicating diverse coefficient distributions), the system switches to individual encoding to achieve better compression ratios, while accepting the trade-off in encoding efficiency only when necessary for optimal compression.
Solution Approach 2:
The encoding parameter (method selection) is changed based on the calculated standard deviation of coefficient magnitudes. This parameter-driven approach allows the system to optimize compression ratio by selecting individual encoding when coefficient diversity is high, while maintaining the ability to use more efficient collective encoding when appropriate.
3Device complexity
If fixed encoding method is used for all wavelet transform coefficients, then device complexity is reduced, but adaptability to different signal types deteriorates
Solution Approach 1:
The patent applies preliminary action by calculating the standard deviation of coefficient magnitudes before selecting the encoding method. This pre-analysis step allows the system to adapt to different signal types without requiring complex real-time adjustments during encoding. The preliminary calculation of statistical parameters enables informed method selection while keeping the overall system relatively simple.
Solution Approach 2:
The system uses feedback from the calculated standard deviation to determine the appropriate encoding method. This feedback mechanism allows the encoding apparatus to adapt to different signal characteristics automatically. The feedback loop (calculate standard deviation → compare with threshold → select encoding method) provides adaptability while maintaining manageable device complexity through a clear decision rule.
Data Source
AI summary
A coefficient encoding unit extracts, as a coefficient group, a group of coefficients (wavelet transform coefficients) located spatially at the same positions from a plurality of high-frequency subbands generated by wavelet transform of a two-dimensional signal, and encodes the group. The coefficient encoding unit has: a unit for collectively encoding the wavelet transform coefficients included in the coefficient group; a unit for individually encoding; a unit for extracting, as a mode determination reference coefficient group, a group of already-encoded wavelet transform coefficients located spatially in neighborhood of the encoding target coefficient group; and a unit for comparing the sum of the absolute values of the wavelet transform coefficients included in the mode determination reference coefficient group with a predetermined threshold, and determining whether to collectively encode or individually encode the wavelet transform coefficients included in the encoding target coefficient group.