Banknote Skew Angle Estimation via FFT Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods struggle to accurately determine the rotational skew angle of banknotes with non-zero angles in high-speed transport systems, especially for torn banknotes lacking regular borders and with more vertically aligned features, as they are difficult for geometric-based algorithms and require calibration, which is not feasible in real-time processing.
Innovation Solution
The method employs a Fast Fourier Transform (FFT) analysis on a subset of the banknote image, using windowing to reduce edge effects and radial line integration to estimate the skew angle based on internal currency design patterns, allowing for fast and accurate skew angle estimation even in arbitrary banknote fragments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If geometric-based algorithms are used to determine skew angle, then the method works for regular documents with clear borders, but it fails for torn banknotes lacking regular borders
Solution Approach 1:
The patent replaces geometric-based algorithms with a frequency-domain approach using Fast Fourier Transform (FFT). Instead of relying on geometric features like borders and lines, the method transforms the image to frequency domain where periodic patterns (such as repeated design elements on banknotes) manifest as peaks. This substitution enables skew angle determination to work on torn banknotes and fragments that lack regular geometric structures, while maintaining high precision through spectral analysis of internal patterns.
2Measurement precision
If calibration procedures are used to determine skew angle, then accuracy can be improved, but real-time processing becomes infeasible
Solution Approach 1:
The patent implements a self-service approach where the banknote itself provides the calibration reference through its internal periodic patterns. The FFT analysis automatically identifies the skew angle by detecting the orientation of frequency peaks corresponding to repeated design elements on the banknote. This eliminates the need for external calibration procedures or reference objects, enabling real-time processing while maintaining high accuracy. The system adapts to each banknote individually without requiring pre-calibration.
3Measurement precision
If methods are designed for text-dominated documents, then horizontal text lines can be detected, but vertically aligned printed features on banknotes create difficulties
Solution Approach 1:
The patent inverts the approach by analyzing vertical frequency components instead of relying on horizontal text lines. The FFT method transforms the image to frequency domain where vertical periodic patterns (common on banknotes) produce horizontal frequency peaks, and horizontal patterns produce vertical peaks. By detecting the orientation of these frequency peaks, the system accurately determines skew angle for banknotes with vertically aligned features, effectively inverting the traditional text-line-based approach to suit banknote characteristics.
Data Source
AI summary
The skew angle of a document image is determined or estimated. The skew angle is determined or estimated by examining patterns found within a Fast Fourier Transform of a subset of the document image. This technique exploits the internal structure of the document designs to allow fast and accurate skew angle estimation, including for arbitrary document fragments.