Rotational Image Checkcode for Machine-Proof User Verification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing checkcode technologies are vulnerable to machine identification due to limited character variations, leading to reduced safety and accuracy in user verification, as either characters are easily enumerable or difficult to identify manually due to excessive deformation.
Innovation Solution
A method and apparatus that pre-store multiple pictures in a database and define rotational directions, where a picture is rotated according to a defined direction upon client request, and the user verifies by inputting the correct rotational direction, enhancing user identification while making machine identification more challenging.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If character variation in checkcode is increased to prevent machine identification, then machine identification difficulty increases, but manual identification difficulty also increases
Solution Approach 1:
The patent transitions from character-based verification to image-based verification with rotational direction identification. Instead of varying character appearances in 2D space, the system adds a rotational dimension, requiring users to identify the correct rotation angle of an image. This dimensional shift increases complexity for machines while remaining intuitive for human users who can easily perceive rotational orientation.
Solution Approach 2:
The system changes the verification parameter from character appearance variation to rotational direction. By storing and verifying the rotational angle parameter rather than character variations, the system achieves higher security against machine identification while maintaining ease of user verification through visual rotational recognition.
2Ease of operation
If character variation is limited to maintain ease of identification, then user verification ease is maintained, but machine identification becomes easier
Solution Approach 1:
The patent introduces rotational direction as a new dimension for verification. Instead of relying on limited character variations, the system uses images rotated to different angles, creating a continuous parameter space for verification that is naturally difficult for machines to enumerate while remaining visually intuitive for users.
Solution Approach 2:
The system replaces character-based textual verification with image-based visual verification. This substitution leverages human visual processing capabilities to easily identify rotational directions while presenting a more challenging problem for automated optical character recognition and image analysis systems.
3Measurement precision
If rotational directions are pre-defined and stored, then verification accuracy increases, but system complexity increases
Solution Approach 1:
The system performs preliminary actions by pre-defining a finite set of rotational directions and storing them in the database. This preparation work is done in advance, allowing the verification process to simply compare the user's input against these pre-established options, thereby maintaining high accuracy without requiring complex real-time processing.
Solution Approach 2:
The system discretizes the rotational parameter into a finite set of predefined directions rather than handling continuous rotation values. This parameter quantization simplifies storage and comparison operations while maintaining sufficient verification accuracy, as the finite set of predefined rotations provides enough distinction to prevent machine enumeration.
Data Source
AI summary
The present disclosure discloses a method, apparatus, and server for user verification to store a plurality of pictures in a database and define a plurality of rotational directions. The method includes: when receiving from a client a request for a checkcode, rotating a picture retrieved from a database according to a defined rotational direction; after storing a correlation between an identification of the client and the rotational direction of the retrieved picture, outputting the rotated picture to the client; receiving a rotational direction of the picture from the client; finding the stored correlation between the identification of the client and the stored rotational direction of the retrieved picture according to the identification of the client; determining whether or not the rotational direction of the picture returned from the client matches the stored rotational direction. If they match, user verification is passed; otherwise, the user verification is failed. The large number of pictures in the database makes it difficult for enumeration. In addition, by using the rotational direction as identification information to verify user, the safety and accuracy of user verification can be enhanced.