Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{193741,
        author = {E D Pavan Kumar and K. Namitha and Sreepati Lakshmi Susritha and Kalasapati Kaveri and Etlam  Muni  Puneeth  Reddy},
        title = {Bank Cheque Verification Using Deep Learning And Image Processing},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {10},
        pages = {1920-1928},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=193741},
        abstract = {This innovative system transforms bank check verification by integrating deep learning, image processing, and a user-friendly Django-based web interface, streamlining the cheque truncation process with minimal human intervention. Leveraging the IDRBT cheque dataset, our convolutional neural network (CNN), implemented with PyTorch, achieves 99.14% accuracy in recognizing handwritten digits, as demonstrated in the base paper, while the source code employs adaptive thresholding and Gaussian blurring for robust image preprocessing. MATLAB’s optical character recognition (OCR), with 97.7% accuracy, extracts machine-printed text such as IFSC codes and account numbers, complemented by Pytesseract in the code for region-based text extraction. Signature verification, powered by Scale Invariant Feature Transform (SIFT) and Support Vector Machine (SVM), attains 98.1% accuracy, with the code implementing SIFT feature extraction and SVM classification for real-time authenticity checks. The web interface enables users to upload cheque images, view datasets, train models, and receive instant classification results (“Genuine” or “Not Genuine”), enhancing accessibility. The system extracts critical details like cheque numbers, amounts, and signatures, adhering to CTS-2010 standards for Indian banks while supporting international formats. By automating verification, it reduces processing time, operational costs, and fraud risks, using contour detection and region-based analysis for precision. This scalable solution, combining the paper’s rigorous methodology with the code’s practical implementation, sets a new standard for secure, efficient financial transactions, with potential for multilingual and multi-format expansions.},
        keywords = {bank cheque verification, deep learning, convolutional neural networks, CNN, optical character recignition, signature verification, image processing, text extraction, support vector machine},
        month = {March},
        }