Image Encoding Apparatus Reducing Memory Cost via Predictive Parameter Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image data encoding methods, such as JPEG-LS, face inefficiencies in parameter updating and memory usage, particularly when determining encoding parameters for predictive errors, and require additional memory for coefficient buffers.
Innovation Solution
An image encoding apparatus that performs wavelet transform and predictive coding in raster scan order, using a symbol generator, encoder, estimation unit, and determination unit to efficiently encode image data while minimizing memory costs by updating encoding parameters based on code lengths and predicting symbols for subsequent coefficients.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a one-line coefficient buffer is used for each subband component to determine encoding parameters, then the accuracy of parameter determination is improved, but the memory cost increases
Solution Approach 1:
The patent extracts only the necessary information (code lengths of previously encoded coefficients) from the buffer, rather than storing entire coefficient lines. This selective extraction maintains parameter determination accuracy while significantly reducing memory requirements compared to buffering complete coefficient lines.
Solution Approach 2:
The patent performs preliminary encoding of coefficients in raster scan order, storing their code lengths in advance. This allows the determination of encoding parameters for subsequent coefficients based on previously computed code lengths, eliminating the need to buffer entire coefficient lines while maintaining accuracy.
2Quantity of substance
If encoding parameters are determined using only the code length of the most recently generated code, then the memory cost is reduced, but the accuracy of parameter update becomes insufficient
Solution Approach 1:
The patent performs preliminary encoding of coefficients in raster scan order, storing their code lengths in advance. This allows the determination of encoding parameters for subsequent coefficients based on previously computed code lengths, eliminating the need to buffer entire coefficient lines while maintaining accuracy.
Solution Approach 2:
The patent uses the code lengths of previously encoded coefficients as feedback to determine encoding parameters for subsequent coefficients. This feedback mechanism allows accurate parameter determination without requiring large buffers, as the system continuously adapts based on recent encoding performance.
3Productivity
If wavelet transform and predictive coding are performed with proper state separation, then the encoding efficiency is improved, but the complexity of the encoding process increases
Solution Approach 1:
The patent segments the encoding process into distinct functional units: wavelet transform, predictive coding, and parameter determination. Each unit operates independently with clearly defined inputs and outputs, making the overall complex process more manageable and implementable while maintaining high encoding efficiency.
Solution Approach 2:
The patent performs preliminary encoding of coefficients in raster scan order, storing their code lengths in advance. This allows subsequent parameter determination to be based on pre-computed data, reducing the real-time processing complexity while maintaining encoding efficiency.
Data Source
AI summary
An encoding apparatus includes a transformer which performs wavelet transform on image data to generate coefficients of subbands, and a predictive encoder which performs predictive coding on the coefficients. The predictive encoder includes a symbol generator which generates a symbol as an encoding target from a predictive error, an encoder which performs entropy coding on the generated symbol using a parameter determined in a process of encoding an immediately preceding coefficient of the coefficient of interest, an estimation unit which estimates a tentative symbol corresponding to a succeeding coefficient to be encoded next to the coefficient of interest, and a determination unit which determines a parameter for the succeeding coefficient from a code length obtained when assuming that the tentative symbol estimated by the estimation unit is obtained by encoding a parameter used when encoding the symbol of the coefficient of interest.