Image Encoding Error Propagation Control via Block Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image encoding methods are prone to errors during wireless transmission, as a single error in the reference block can propagate across the entire image, leading to significant data loss and reduced resilience.
Innovation Solution
The method involves segmenting images into uniform blocks, applying a frequency-based transform, partitioning blocks into sets and slices, and using prediction based on reference blocks within each slice to constrain errors, while also quantizing and converting coefficients into binary code, with optional down-sampling and pre-filtering to enhance robustness.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If prediction is performed across the whole image or whole image portion, then compression ratio is improved, but error propagation increases
Solution Approach 1:
The image is divided into multiple image portions, and each image portion is further divided into blocks that are processed independently. This segmentation prevents errors from propagating across the entire image, as errors are confined to individual blocks or small groups of blocks within each image portion.
Solution Approach 2:
Different processing strategies are applied to different regions. Specifically, prediction is applied locally within blocks or small groups of blocks rather than globally across the entire image, allowing compression while limiting error propagation scope.
2Reliability
If blocks are partitioned into multiple sets, then error propagation is reduced, but data size increases
Solution Approach 1:
Blocks are divided into multiple sets (e.g., even blocks and odd blocks) that can be processed and transmitted independently. This segmentation allows error confinement while managing data size through efficient encoding of each set.
Solution Approach 2:
Instead of processing all blocks with full prediction and encoding, the method selectively applies processing to different block sets, using prediction for some blocks and alternative methods for others, thereby balancing compression efficiency with error resilience.
Data Source
Figure 1a~1B
Figure 2~6
Figure 7~8
AI summary
A method for encoding data defining an image is disclosed. The image is segmented into image blocks, each image block having a uniform block size. A frequency-based transform to is applied to each of the image blocks, thereby providing transformed image data in which the image data is represented as coefficients defining a linear combination of predetermined basis functions having different spatial frequencies. The blocks of transformed image data are partitioned into one or more sets of blocks, and each set of blocks is further partitioned into a plurality of slices of blocks. Each slice consists of a number of consecutive blocks in the set, and comprises a reference block. The coefficients in subsequent blocks are represented as a prediction based on the coefficients in the reference block. The coefficients are then quantised, and converted into binary code.