Combined Approach to Signature Verification Based on Fourier Transform, Hough Transform, Radon Transform, and the Support Vector Machine Method

This article presents a comprehensive and systematized methodology for applying advanced digital image processing techniques to address the challenges of signature verification. The authors conduct an in-depth investigation of Fourier, Hough, and Radon transform algorithms, which are utilized to extract distinctive and characteristic features from signature images. The study evaluates the contribution of these algorithms to enhancing the accuracy of authentication processes, providing a detailed analysis of their effectiveness in different scenarios. By combining theoretical reasoning with extensive experimental data, the authors develop a robust software platform designed for various stages of the signature verification process. This platform includes modules for training and test sample preparation, image preprocessing, extraction of unique features, and subsequent classification using the support vector method (SVM). The experimental results highlight the high efficiency of the proposed approach, demonstrating its ability to significantly improve the accuracy and reliability of signature verification. The findings suggest that the methodology has strong potential for practical applications in document security and biometric identification systems, where the need for reliable authentication is critical. The article also discusses the limitations of the current approach and outlines potential directions for future research, including the integration of machine learning techniques and the exploration of additional feature extraction methods to further enhance performance.