Image-Based Authentication Using Shape Combinations

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing authentication systems, such as CAPTCHA, are often frustrating for human users while being easily bypassed by sophisticated automated agents, failing to effectively distinguish between human and automated access to computer resources.

Innovation Solution

An authentication challenge system that presents users with image-based challenges, such as selecting objects based on spatial relationships or characteristics, which are difficult for automated processes to emulate but manageable for humans, to determine the type of user accessing computer resources.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If traditional CAPTCHA systems are used for authentication, then automated access can be blocked, but human users experience frustration and difficulty

Engineering Contradiction:
Improveability to block automated accessVSAvoiduser experience for human users
Core Design Contradiction:
ReliabilityVSEase of operation

Solution Approach 1:

The patent transforms the authentication challenge from text-based recognition to spatial relationship recognition. By changing the parameter of what users need to recognize (from characters to spatial arrangements of objects), the system maintains reliability in blocking automated agents while significantly improving ease of operation for human users, who naturally excel at spatial reasoning

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent replaces the mechanical/visual recognition system with a spatial reasoning system. Instead of asking users to visually decode distorted text (mechanical recognition), the system presents spatial relationships between objects that require logical reasoning, thereby maintaining security while improving user experience

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Reliability

If sophisticated authentication challenges are implemented to block automated agents, then security is improved, but the system complexity increases

Engineering Contradiction:
Improvesecurity against automated agentsVSAvoidsystem complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent segments the authentication challenge into distinct spatial relationships (above, below, left, right) and separate object identifiers. This segmentation allows the system to present complex spatial reasoning tasks while maintaining a simple, clean interface and straightforward validation logic, thereby improving security without proportionally increasing system complexity

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Instead of presenting users with a complex task and providing simple validation, the patent inverts the approach by presenting simple spatial relationship questions that are easy to validate but difficult for automated agents to solve. The complexity lies in the cognitive task rather than the system implementation

Inventive Principle:
Principle #13The other way round (Inversion)

Data Source

PatentUS20230297661A1Computer challenge systems based on shape combinations
Publication Date: 2023.09.21 ARKOSE LABS HOLDINGS INC
  • US20230297661A1 patent drawing
  • US20230297661A1 patent drawing
  • US20230297661A1 patent drawing

AI summary

A method includes sending a challenge data structure to a user computer system. The challenge data structure defines a challenge user interface to be presented to a user of the user computer system. The challenge user interface includes a key shape and a key number, and prompts the user to select a correct image from a plurality of images. Each of the images includes a combination of shapes, and the correct image comprises a first shape of the shapes that corresponds to the key shape and has a same number of the first shape as the key number. The method includes obtaining a user input to the challenge user interface that represents a selection of at least one image from the images, and providing access to the computer resource for the user computer system based on whether the at least one image is consistent with the correct image.