2-D Symbol Feature Representation for Image Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current machine learning methods lack efficient ways to represent and classify features from objects and natural language texts, particularly in a format that leverages image processing techniques effectively.
Innovation Solution
A two-dimensional (2-D) symbol is created using a matrix of N×N pixels, partitioned into sub-matrices, where each sub-matrix contains a feature, with symbol creation rules determining importance order, size, and location, allowing for image processing techniques like CNNs to classify the meaning of combined features.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If features are represented in traditional machine learning formats, then classification can be performed using standard algorithms, but the ability to leverage image processing techniques like CNNs is lost
Solution Approach 1:
The patent transforms feature data from traditional tabular or vector formats into a two-dimensional spatial representation. Features are organized in a grid structure where rows and columns represent different feature dimensions, enabling the application of image processing techniques like convolutional neural networks that operate on 2D data structures. This dimensional transformation allows standard machine learning features to be processed by vision-based algorithms.
2Measurement precision
If features are transformed into 2-D symbol format for image processing, then classification accuracy is enhanced, but the complexity of feature representation and symbol creation increases
Solution Approach 1:
The patent segments the feature representation into discrete sub-matrices, where each sub-matrix corresponds to a specific feature or group of features. This segmentation allows for systematic organization of features in the 2-D symbol, making the creation process more manageable. Each sub-matrix can be independently constructed and then assembled into the complete 2-D symbol structure.
Solution Approach 2:
The patent transforms feature parameters into visual parameters suitable for image processing. Numerical features are converted into visual patterns, colors, or textures within sub-matrices, while categorical features are represented through distinct visual symbols. This parameter transformation enables the 2-D symbol to preserve the semantic meaning of original features while becoming compatible with image processing algorithms.
3Reliability
If all features are included in the 2-D symbol, then comprehensive classification is achieved, but the size and processing time of the symbol increases
Solution Approach 1:
The patent extracts and separates individual features or feature groups into distinct sub-matrices within the 2-D symbol. This extraction allows for selective processing where not all sub-matrices need to be analyzed with the same computational intensity. Important features can be highlighted or given larger sub-matrices, while less critical features use smaller representations, optimizing processing time while maintaining classification completeness.
Data Source
AI summary
Methods of facilitating machine learning via a 2-D symbol are disclosed. Features of an object are received in a first computing system having a 2-D symbol creation application module installed thereon. A multi-layer 2-D symbol is formed from the features according to a set of symbol creation rules. 2-D symbol is a matrix of N×N pixels partitioned into a number of sub-matrices with each sub-matrix containing one feature, where N is a positive integer. Meaning of the combined features in the 2-D symbol is learned in a second computing system by using an image processing technique to classify the 2-D symbol transmitted from the first computing system. The symbol creation rules determine the importance order, size and location of sub-matrices in the 2-D symbol.