Morphologically-Adaptive Coding Networks for Queryable Data Compression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data encoding techniques, such as JPEG, lose order and meaning in the compressed data stream, requiring reconstruction into raw data for manipulation, which compromises data quality and efficiency.
Innovation Solution
The development of morphologically-adaptive coding networks with layers of nodes and connections based on energy dissipation, allowing for efficient encoding and decoding of data while retaining structure and quality, using a biomimetic approach inspired by biological neural networks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If existing encoding techniques (e.g., JPEG) are used to compress data, then data size is reduced for efficient transmission and storage, but order and meaning are lost in the compressed data stream
Solution Approach 1:
The patent segments the compressed data stream into meaningful units by tracking visited nodes and their hierarchical relationships in the coding network. Each compressed data element contains metadata about its position in the original data structure, allowing segmentation that preserves semantic meaning while reducing data size.
Solution Approach 2:
The patent introduces an intermediary indexing structure that maps compressed data elements back to their original positions and relationships. This intermediary layer allows the compressed data to retain order and meaning without requiring full reconstruction, serving as a bridge between the compressed and original data states.
2Ease of operation
If compressed data is manipulated by reconstructing into raw data, then manipulation can be performed, but data quality degrades and processing efficiency decreases
Solution Approach 1:
The patent performs preliminary actions by pre-computing and storing the hierarchical relationships and node visitation paths during the compression phase. This allows manipulation operations to be performed directly on the compressed data using the pre-computed structure, avoiding the need for time-consuming reconstruction while maintaining manipulation capability.
Solution Approach 2:
The patent replaces the mechanical reconstruction process with a computational indexing system. Instead of physically reconstructing the raw data structure to enable manipulation, the system uses computational indexes and metadata to allow direct operations on compressed data, significantly improving processing efficiency.
3Quantity of substance
If existing encoding techniques are used, then compression is achieved, but the compressed data stream represents random data requiring full reconstruction
Solution Approach 1:
The patent introduces an intermediary indexing structure that preserves the relationship between compressed data elements and their original positions. This intermediary layer allows selective access and manipulation without full reconstruction, reducing the complexity of data processing operations while maintaining compression benefits.
Solution Approach 2:
The patent adds a new dimension to the compressed data by incorporating metadata about node visitation paths and hierarchical relationships. This additional dimension allows the compressed data to be navigated and manipulated directly without reconstruction, transforming the data structure from a flat random stream to a multi-dimensional navigable structure.
Data Source
AI summary
Examples of biomimetic codecs and biomimetic coding techniques are described herein. Morphologically-adaptive coding networks can be developed in accordance with energy dissipation driven “heat” generated by application of training data. The morphologically-adaptive coding networks may be representative of common features expected in an input signal or data stream. Decoding may proceed using the morphologically-adaptive coding network. Morphologically-adaptive coding networks may be used as a cortex that can be shared for boosting multimedia data compression rates and/or increasing the encode-decode fidelity of information content while the features remain queryable in encoded form. Examples of the biomimetic codecs and biomimetic coding techniques provide a broad-based technology platform that can be used in context-IDed multimedia storage, pattern recognition, and high-performance computing/big data management, the hallmarks of web- and cloud-based systems.