Eyeball Gaze Tracking for Human-Machine Verification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current human-machine verification methods are inadequate in distinguishing between real-person and machine attacks, as they can be bypassed by simulation methods, leading to increased costs for machine attacks and reduced user experience due to higher verification difficulty for real persons.
Innovation Solution
A human-machine verification method that collects and analyzes eyeball gaze point tracks using a classification model trained on historical behavior similarities to determine whether a user is a real person or a machine, thereby improving verification reliability and user experience by automating the verification process without requiring user awareness.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If user-behavior interactive verification is adopted to identify real person access and machine attack, then machine attack cost is increased, but verification difficulty for real persons is increased, thus user experience is reduced
Solution Approach 1:
The patent replaces mechanical interaction verification (clicking, sliding) with optical field verification by using eyeball gaze point tracking. The eyeball movement tracking system captures and analyzes gaze points without requiring conscious user participation, thus maintaining verification reliability while eliminating the need for complex user operations.
Solution Approach 2:
The system uses the user's own eyeball movements as the verification mechanism. The eyeball gaze points naturally reflect user intent and behavior patterns without requiring the user to perform additional actions. This self-service approach maintains high reliability while keeping the verification process invisible and effortless for real users.
2Reliability
If verification difficulty is increased continuously to increase machine attack cost, then machine attack cost is increased, but user experience is reduced due to increased difficulty for real persons
Solution Approach 1:
The patent substitutes complex mechanical interaction verification with optical gaze tracking. Instead of requiring users to complete complex tasks like identifying objects or following intricate paths, the system passively tracks eyeball movements which naturally reveal user intent. This reduces verification process complexity while maintaining security through sophisticated gaze pattern analysis.
Solution Approach 2:
The system changes the verification parameter from active behavior (clicks, slides) to passive physiological parameter (eyeball gaze points). By analyzing gaze point coordinates, movement speed, and停留 time, the system achieves high-security verification without requiring complex user interactions, thus reducing overall verification complexity.
3Reliability
If behavior simulation methods are developed to bypass user-behavior interactive verification, then verification reliability is reduced, but machine attack cost is decreased
Solution Approach 1:
By replacing mechanical interaction verification with optical gaze tracking, the patent creates a verification mechanism that is difficult to simulate. While behavior simulation can replicate clicks and slides, simulating realistic eyeball movement patterns including saccades, fixations, and smooth pursuits requires sophisticated hardware and software that increases attack implementation difficulty and cost.
Solution Approach 2:
The patent introduces eyeball gaze points as an intermediary verification layer between user identity and system access. This intermediary physiological parameter provides an additional security layer that is difficult to forge, as it requires both hardware (eye tracking device) and software coordination, thereby maintaining verification accuracy while increasing attack complexity.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The present application discloses a human-machine verification method and apparatus, a device and a storage medium, and relates to the fields of Internet security technologies and computer vision technologies. An implementation includes: receiving an identity verification request sent by a requester, and collecting an eyeball gaze point track on an identity verification page, the identity verification request including identity verification information; identifying whether the identity verification information is correct based on pre-stored user identity information; if the identity verification information is correct, performing classification based on the eyeball gaze point track on the identity verification page using a first classification model, and outputting a first probability value indicating whether the requester is a real person or a machine; and determining based on the first probability value that the requester is the real person or the machine, and outputting an identity verification result indicating that identity verification is passed or fails. With the present application, a reliability of human-machine verification and user experiences may be improved at the same time.