Automated Anti-Counterfeiting Evaluation Using Machine Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current anti-counterfeiting measures are often ineffective due to reliance on intuition and limited authentication protocols, which fail to accurately assess the effectiveness of combinations of overt and covert measures, leading to compromised authentication processes.
Innovation Solution
An assessment and evaluation system that employs machine learning techniques, such as autoencoders and convolutional neural networks, to derive features from items and evaluate the effectiveness of anti-counterfeiting measures by generating difference maps and effectiveness metrics, using techniques like 'hot spot' and clustering evaluation methods.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If product designers rely on intuition when applying anti-counterfeiting measures, then the design process is simple and quick, but the effectiveness of the anti-counterfeiting measures is not well understood and may be inaccurate
Solution Approach 1:
The patent replaces the mechanical system of human intuition and visual inspection with an automated machine learning system that uses algorithms to evaluate anti-counterfeiting measures. The system automatically analyzes images of products with anti-counterfeiting measures and provides effectiveness evaluations, substituting human judgment with computational analysis to improve measurement precision while maintaining ease of use.
2Ease of operation
If consumers focus on accurately duplicated overt measures during authentication, then the authentication process is simple, but covert measures may not be effective as consumers do not consider them
Solution Approach 1:
The patent introduces feedback by using machine learning models to analyze both overt and covert anti-counterfeiting measures and provide effectiveness evaluations. The system processes images of products and generates feedback about the effectiveness of different measures, including covert ones, helping designers understand which measures are actually effective even if consumers don't actively check them.
Solution Approach 2:
The patent introduces an intermediary system (the automated evaluation platform) that mediates between the anti-counterfeiting measures and the authentication process. This intermediary automatically evaluates the effectiveness of covert measures without requiring consumers to actively search for them, while still providing information about overall authentication effectiveness.
3Reliability
If many covert measures are included in a product, then the anti-counterfeiting capability is enhanced, but consumers may focus on them as a whole and not on each one individually, reducing effectiveness
Solution Approach 1:
The patent applies segmentation by breaking down the evaluation of anti-counterfeiting measures into individual components. The machine learning system analyzes different types of measures (overt, covert, physical, digital) separately and provides effectiveness evaluations for each type, allowing for precise assessment of individual measures even when multiple are present on a single product.
Data Source
AI summary
A system for evaluating the effectiveness of an anti-counterfeiting measure employed for an item is provided. The system trains a classifier to indicate whether the anti-counterfeiting measure of an evaluation item is genuine or counterfeit. For evaluation items that have been classified as genuine or counterfeit, the system applies the classifier to determine whether the anti-counterfeiting measure of that evaluation item is genuine or counterfeit. The system then generates an effectiveness metric that indicates whether the anti-counterfeiting measure is effective based on evaluation items that are assessed as being genuine whose anti-counterfeiting measures are classified as being counterfeit.