Automated Signature Verification in Bank Cheque Processing Using Siamese and Convolutional Neural Networks
Disha Tuteja, Raman Dhand, Mrigya Reineu, Jyotika Israni, Divyanshu Singh, Khushi Dhal · 2024
Bank cheques play a pivotal role in global financial transactions, with a vast number of them, processed daily. The conventional manual verification process, involving visual scrutiny of crucial details, is both time-consuming and labour-intensive, often leading to delays in transaction validation. This paper proposes an automated signature verification method for bank cheques, leveraging advancements in image processing and machine learning. Our approach begins with capturing a customer's handwritten signatures from cheque leaves. Our method outlines the various processing steps, including grayscale binarization, signature detection using OCR and Linesweep, and classification using SNN.