Marker-less Pre-paid Card Deal Engine Using Image Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Pre-paid transaction cards lack the ability to provide personalized and exciting deal recommendations, failing to leverage user data and transaction histories effectively, especially in gifting scenarios.
Innovation Solution
A transaction system that includes a deal engine communicatively coupled with a transaction database, capable of receiving images of marker-less portions of pre-paid transaction cards, deriving image descriptors, and constructing personalized deal recommendations based on transaction characteristics, which are then transmitted to users as augmented real-time images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If pre-paid transaction cards use traditional marker-based augmented reality systems, then the system can track the position of the card, but the system is limited to only those transaction cards that have markers and fails to leverage user data for personalized recommendations
Solution Approach 1:
The patent extracts the marker component from the augmented reality system, transitioning from marker-based to marker-less image recognition. This allows the system to work with any transaction card regardless of whether it has markers, while simultaneously enabling access to card information through image analysis for personalized recommendations
Solution Approach 2:
The image recognition system serves multiple functions: it identifies the card type, extracts card information, enables augmented reality display, and provides the basis for personalized recommendations. This multi-functional approach resolves the contradiction by making the system universally applicable while充分利用 user data
2Ease of operation
If pre-paid transaction cards provide basic transaction functionality, then the cards are simple to use, but they fail to provide personalized and exciting deal recommendations
Solution Approach 1:
The system automatically analyzes transaction history and user preferences without requiring user input, generating personalized recommendations autonomously. This maintains ease of operation while fully utilizing transactional data to provide customized deal suggestions
Solution Approach 2:
The system continuously learns from user interactions with recommendations, refining future suggestions based on user behavior patterns. This feedback mechanism enables personalized recommendations while keeping the interface simple and easy to use
3Measurement precision
If the system uses marker-based tracking for augmented reality display, then the position of the card can be tracked, but the system requires pre-designated markers on the cards
Solution Approach 1:
The patent replaces the mechanical marker-based tracking system with an optical image recognition system. This substitution eliminates the need for physical markers on cards while maintaining accurate position tracking through advanced image processing algorithms
Solution Approach 2:
The system changes the recognition parameter from detecting specific marker patterns to analyzing general card image features. This parameter change allows position tracking without requiring special markers, reducing card design complexity while maintaining tracking precision
Data Source
AI summary
Pre-paid transaction card systems and methods are described. A transaction system can include a transaction database to store transaction objects that represent characteristics of a pre-paid transaction card. A deal engine communicatively coupled with the transaction database can receive an image of a marker-less portion of the transaction card and derive a set of image descriptors from the marker-less portion. The deal engine can also identify transaction characteristics associated with the card from the transaction database using the image descriptors. In addition, the deal engine can construct a deal recommendation based on the transaction characteristics previously identified. The deal recommendation can be transmitted to a user, and in some cases, displayed to the user as an augmented reality image.