Image Encoding with Slice-Based Prediction for Error Containment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image encoding methods suffer from propagation of errors across the entire image due to prediction based on a single reference block, leading to reduced resilience in data transmission, especially in wireless environments.
Innovation Solution
The method segments images into uniform blocks, applies frequency-based transforms, partitions blocks into slices with a reference block, and uses prediction within each slice to constrain errors, followed by quantization and binary encoding, with optional grouping and interleaving to enhance resilience.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of substance
If prediction is performed across the whole image or whole image portion using a single reference block, then compression ratio is improved, but error propagation increases
Solution Approach 1:
The image is divided into multiple slices, with each slice having its own reference block. Prediction is performed independently within each slice rather than across the entire image. This segmentation confines error propagation to individual slices, preventing a single error from corrupting the whole image, while still maintaining compression through prediction within each slice.
2Reliability
If blocks are grouped into sets with non-sharing boundaries, then error propagation is reduced, but device complexity increases
Solution Approach 1:
Blocks are organized into multiple sets where blocks in any one set do not share boundaries with other blocks in the same set. This creates independent processing units that isolate errors. The non-sharing boundary property ensures that errors in one block do not propagate to adjacent blocks in the same set, enhancing error containment while maintaining manageable processing complexity through systematic organization.
Data Source
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.