Image Decoding Device Non-Zero Coefficient Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
High bandwidth requirements between embedded processors and external storage devices in mobile devices hinder the decoding speed of high-definition video images, particularly due to frequent data reading and writing needed for video image/picture decoding.
Innovation Solution
An image decoding method and device that performs entropy decoding on blocks of an input bitstream, storing only the block position information and coefficient values of non-zero coefficients, thereby reducing data storage and retrieval from external storage, and utilizing auxiliary messages for subsequent reconstruction procedures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all decoded coefficients are stored to external storage device, then complete image data is preserved, but bandwidth requirement and storage overhead increase significantly
Solution Approach 1:
The patent extracts only the essential non-zero coefficients from the decoded data for storage in external memory, while keeping zero coefficients in the processor's buffer. This selective extraction reduces the data volume written to external storage while preserving all necessary information for image reconstruction, thereby resolving the contradiction between data completeness and storage bandwidth requirements.
2Reliability
If frequent data reading/writing is performed on external storage device, then decoding completeness is maintained, but decoding speed decreases due to bandwidth bottleneck
Solution Approach 1:
The patent applies extraction by writing only non-zero coefficients to external storage, significantly reducing the frequency and volume of I/O operations. This allows the system to maintain decoding completeness while avoiding the bandwidth bottleneck that would otherwise limit decoding speed in high-definition video processing.
3Ease of manufacture
If zero coefficients are stored to external storage device, then storage is simplified, but bandwidth is wasted on storing redundant data
Solution Approach 1:
The patent changes the storage parameter by selectively writing only non-zero coefficients to external storage based on their magnitude values. This parameter-based filtering eliminates redundant zero coefficient storage operations, optimizing bandwidth utilization while maintaining storage simplicity for the essential data.
Data Source
AI summary
Image decoding methods are provided. First, an input bitstream corresponding to an image is acquired and an entropy decoding operation is performed on a block to be decoded in the input bitstream to obtain a decoding result, wherein the decoding result includes at least one coefficient having a corresponding block position information and coefficient value. It is then determined whether the coefficient value of the at least one coefficient is zero. When the coefficient value of the at least one coefficient is not zero, the at least one coefficient is determined as an non-zero coefficient and the block position information and coefficient value corresponding to the non-zero coefficient are stored to an external storage device. When the coefficient value of the at least one coefficient is zero, the block position information and coefficient value corresponding to the at least one coefficient are not stored.