Vector-Based Character Encryption via ML Noise Obfuscation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The complexity and vulnerability of public and private key systems in encrypting and decrypting data make them susceptible to hacking, as key distribution adds complexity and creates points of vulnerability.
Innovation Solution
A system that encrypts data by splitting characters into vector-based characters, rotating and organizing them, and adding noise vectors using a machine learning model to obscure the original text, making it difficult for hackers to decrypt, while decryption involves removing noise vectors to recover the original data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If public and private key systems are used to encrypt data, then data security is improved, but system complexity and vulnerability increase
Solution Approach 1:
The patent extracts the key management complexity from the encryption system by using visual cryptography where the encryption key is a simple visual indicator (such as a color or shape) rather than complex cryptographic keys. This separates the security function from the complex key management infrastructure.
Solution Approach 2:
The patent replaces the mechanical/key-based cryptographic system with a visual/image-based system. Instead of using mathematical key pairs, the system uses visual transformations of images that can be encrypted and decrypted through visual processing rather than cryptographic key exchange.
2Reliability
If public and private key systems are used to encrypt data, then data security is improved, but vulnerability to hacking increases
Solution Approach 1:
The patent converts the vulnerability of key distribution into a benefit by eliminating the need for key distribution entirely. The visual cryptography system allows secure sharing through image transformation without requiring secure key exchange channels, turning the previously harmful key distribution process into a benign visual sharing process.
Solution Approach 2:
The patent introduces visual image transformation as an intermediary between the plaintext and ciphertext. Instead of directly encrypting data with cryptographic keys, the system uses visual transformations (such as filtering, color manipulation, or geometric transformations) as an intermediary layer that provides security without exposing cryptographic keys.
3Reliability
If characters are split into vector-based representations and noise is added, then encryption strength is improved, but processing complexity increases
Solution Approach 1:
The patent segments characters into vector-based representations and further divides them into component parts (such as strokes, curves, or geometric elements). This segmentation allows for granular manipulation of individual character components and enables the addition of noise at the vector level, strengthening encryption while maintaining structured processing.
Solution Approach 2:
The patent changes parameters of the vector-based character representations, such as modifying line thickness, curvature, intersection points, or spatial relationships. By altering these parameters and adding controlled noise, the system strengthens encryption while the structured nature of vector graphics allows for relatively efficient processing compared to pixel-based approaches.
Data Source
AI summary
In an example embodiment, a solution is provided for encrypting and decrypting data in which the solution itself creates unique symbols, reducing or eliminating the possibility that a hacker or other malicious actor can understand what the symbols mean, let alone decrypt them. More particularly, for encryption, the characters of an original text is split into individual vector-based characters, and each of these vector-based characters are split into subcharacters at intersection points. Each of the split characters are then rotated, and the rotated characters are organized one on top of each other. The characters are then connected to each other, and the intersection points of the subcharacters within the characters that were used to split the characters into subcharacters are passed into a machine learning model that is trained to add lines between intersection points that have no lines, as noise to further obfuscate the original text.