Multispectral CNN Analysis for Security Document Verification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for analyzing security documents are limited in their ability to effectively utilize infrared and ultraviolet information, which can contain critical security elements and tampering evidence, beyond visual inspection.
Innovation Solution
A computer-implemented method and system that incorporates convolutional neural networks (CNNs) to analyze both visible-color data and infrared/ultraviolet data from security documents, allowing for the extraction of characteristics and enhanced security checks by treating IR and UV data as additional input layers, enabling extended automated optical analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If only visible-color data is fed to the CNN for analysis, then the device complexity remains low, but the measurement precision and reliability of security document verification deteriorate because infrared and ultraviolet security elements cannot be detected
Solution Approach 1:
The CNN is extended to process multiple types of input data (visible-color, infrared, and ultraviolet) through a unified architecture. The network accepts multi-spectral data as additional input channels or layers, allowing a single model to perform both visible and non-visible spectrum analysis, thereby improving detection accuracy without requiring separate processing systems for each spectral range
Solution Approach 2:
The patent extends the input space of the CNN by adding spectral dimensions beyond the traditional three-color RGB input. By incorporating infrared and ultraviolet data as additional input layers or channels, the network operates in a higher-dimensional feature space, enabling it to detect security elements that are invisible in the standard visible spectrum while maintaining the existing CNN processing framework
2Reliability
If multiple separate analysis systems are used for visible, infrared, and ultraviolet data, then the measurement precision improves through comprehensive security element detection, but the device complexity and computational requirements increase significantly
Solution Approach 1:
The patent merges the processing of visible-color, infrared, and ultraviolet data into a single CNN architecture. By combining multi-spectral inputs and processing them through shared convolutional layers and pooling operations, the system achieves comprehensive security verification reliability while avoiding the complexity of maintaining and coordinating multiple separate analysis systems
Solution Approach 2:
The extended CNN serves as a universal processing system that handles all spectral ranges (visible, infrared, and ultraviolet) through a single model. This multi-functional approach improves verification reliability by comprehensively analyzing all security elements across different spectra while reducing device complexity compared to using separate specialized systems for each spectral range
Data Source
AI summary
The invention refers to a computer-implemented method of analyzing security documents having visible information and at least one of infrared and ultraviolet detectable information, the method comprising the steps of: receiving visible-color data of a first set of pixels of a first region of the security document in a first image of the security document and feeding the visible-color data to a convolutional neural network, CNN; receiving infrared, IR, and/or ultraviolet, UV, data of a second and/or third set of pixels of a respective second and/or third region in a respective second and/or third image of the security document and feeding the IR and/or UV data to the CNN; analyzing the visible-color data of the first set of pixels using the CNN to extract characteristics of the security document from the visible information; and analyzing the IR and/or UV data of the second and/or third set of pixels using the CNN to extract characteristics of the security document from the IR and/or UV information. Further, a corresponding system is provided.