Machine Vision Authentication Using Intrinsic Material Patterns
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for authenticating and identifying physical objects face challenges due to variations in materials, camera angles, lighting conditions, and environmental factors, leading to unreliable image matching and increased counterfeiting, especially with removable identifiers that fade or are easily duplicated.
Innovation Solution
A machine vision agnostic protocol using computer vision and machine learning algorithms to generate a tamper-proof physical code based on an object's distinctive material structure, capturing images under varied conditions and creating a machine vision agnostic protocol to authenticate or identify objects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional identifiers (barcodes, holographic stickers) are physically attached to objects for authentication, then identification functionality is achieved, but the identifiers fade with usage, are easily duplicated, or can be removed
Solution Approach 1:
The patent extracts the authentication information from physical identifiers and embeds it directly into the object's material structure through manufacturing process footprints. This eliminates the need for separate physical identifiers that can fade or be removed, as the authentication data becomes an intrinsic property of the object itself.
Solution Approach 2:
The patent introduces machine vision algorithms and microstructural feature analysis as intermediaries between the object and the authentication system. Instead of directly reading physical identifiers, the system captures images and analyzes microstructural patterns, creating a reliable bridge that overcomes the limitations of traditional identifiers.
2Adaptability or versatility
If image matching is used to authenticate objects under various conditions, then identification capability is provided, but variations in lighting, camera angles, and materials reduce matching accuracy
Solution Approach 1:
The patent performs preliminary analysis of microstructural features during the authentication setup phase. By pre-identifying and storing reference microstructural patterns under various conditions, the system prepares in advance for different authentication scenarios, enabling accurate matching despite variations in lighting, camera angles, or materials.
Solution Approach 2:
The patent changes the parameters used for authentication from surface-level visual features to deep microstructural characteristics. By analyzing fundamental material properties rather than superficial appearance, the system maintains high precision across varying external conditions such as lighting and camera specifications.
3Ease of operation
If removable identifiers are used on products, then identification is enabled, but the identifiers must be removed for product use and are vulnerable to counterfeiting
Solution Approach 1:
The patent merges the authentication identifier with the object's material structure itself. The manufacturing process footprints become both the functional material and the authentication code simultaneously, eliminating the need for separate removable identifiers and making counterfeiting extremely difficult without access to the original manufacturing process.
4Reliability
If physical identifiers are permanently attached to objects, then tracking capability is maintained, but critical items cannot be tagged and identifiers are easy to duplicate
Solution Approach 1:
The patent enables objects to authenticate themselves through their inherent microstructural features. The manufacturing process automatically embeds unique identification patterns during production, and the authentication system simply needs to read these self-contained features, eliminating the need for complex external tagging systems while maintaining high tracking reliability.
Data Source
AI summary
The present disclosure relates to a method for creating a machine vision agnostic protocol for authenticating physical objects using computer vision (CV) and machine learning (ML) algorithms. Multiple images of a physical object are captured in various conditions, including different materials, shapes, cameras, lighting, and alignments. These images form datasets, from which a focus area is auto-selected using ML algorithm based on the object's features. The machine vision agnostic protocol identifies objects using unique similar patterns from the focus areas. To authenticate an image of a query object captured using a camera, multiple images of the query focus area are taken, and the best quality image is selected. Feature vectors are created and compressed into a query physical code using data from the machine vision agnostic protocol. The query physical code is compared with a pre-stored physical code to determine the object's authenticity.