3D Object Manipulation for Bot Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current CAPTCHA systems are becoming less effective due to advancements in OCR technology, allowing robots to recognize alphanumeric sequences, leading to increased robotic attacks on sensitive websites, and struggle to distinguish human users from robots in a timely manner.

Innovation Solution

A 3D object manipulation system is implemented, where users must recreate a pattern displayed on 3D elements by manipulating their position and orientation, preventing robots from accessing websites and allowing human users to access securely within a limited time frame.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If CAPTCHA uses traditional alphanumeric sequences, then robots can successfully recognize and bypass them using OCR technology, but the system fails to effectively distinguish human users from robots

Engineering Contradiction:
Improvebot detection accuracyVSAvoidrobotic attack success rate
Core Design Contradiction:
ReliabilityVSExtent of automation

Solution Approach 1:

The patent transitions from 2D flat images to 3D objects with multiple faces that can be rotated and manipulated in space. Users must manipulate 3D elements to reconstruct a target pattern, adding spatial dimensionality that defeats traditional OCR while remaining solvable by humans

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Solution Approach 2:

The patent introduces dynamic manipulation where 3D elements can be rotated, moved, and repositioned by users. This dynamic interaction requires human motor skills and spatial reasoning, making automated recognition impossible while keeping the challenge accessible to humans

Inventive Principle:
Principle #15Dynamics

2Object-affected harmful factors

If the system uses complex 3D manipulation tasks, then robot access is effectively blocked, but the test becomes difficult for human users to complete in limited time

Engineering Contradiction:
Improverobotic attack preventionVSAvoiduser completion time
Core Design Contradiction:
Object-affected harmful factorsVSLoss of time

Solution Approach 1:

The patent divides the target pattern into multiple separate 3D elements (cubes, spheres, pyramids) that users must manipulate individually. This segmentation makes the task manageable for humans while preventing robots from processing the entire pattern simultaneously

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent adjusts parameters such as the number of 3D elements, their complexity, and the manipulation requirements to optimize the balance between security and user experience, ensuring the test is solvable by humans but intractable for automated systems

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP2750064B13D bot detection
Publication Date: 2020.02.12 DASSAULT SYSTEMES SA
  • EP2750064B1 patent drawingFigure 1
  • EP2750064B1 patent drawingFigure 2
  • EP2750064B1 patent drawingFigure 3

AI summary

In one embodiment, a computer method of verifying an operator is human includes automatically selecting a pattern and dividing the pattern into a plurality of textures. The method further includes projecting each texture onto a different respective displayed element in a 3D experience. The method additionally includes randomizing a position and/or an orientation of at least one displayed element of the different respective display elements in the 3D experience. The method also includes receiving operator manipulations of the randomized elements to solve for the pattern and make the pattern appear/reappear. The method further includes granting access to the operator if the pattern is made to appear/reappear. Access is granted upon determining that the operator has made the pattern is appear/reappear, and access is denied if the operator fails to make the pattern appear/reappear.