Signature Fraud Detection using Deep Learning
Deep Learning Approach to Offline Signature Forgery Prevention
IEEE BASE PAPER ABSTRACT:
Image analytics solutions are very crucial for not only security, but also for authentication purposes. One such intriguing new Deep Learning application is automatic signature verification. Signed papers are the foundation of all transactions and governance in banks and government agencies, to mention a few examples. Verifying signatures on these documents is an important aspect of the back-office operations. Manual signature verification is not only time intensive, but also prone to mistakes and possible fraud. These flaws may be removed using AI-based automated signature analysis, resulting in considerable efficiency benefits. Based on the Siamese Neural Network, this research proposes an offline solution for signature identification and verification. To construct a feature vector, a collection of global features is utilised.
PROJECT OUTPUT VIDEO:
ALGORITHM / MODEL USED:
OUR PROPOSED PROJECT ABSTRACT:
SYSTEM REQUIREMENTS:
HARDWARE REQUIREMENTS:
- System : Pentium i3 Processor.
- Hard Disk : 500 GB.
- Monitor : 15’’ LED.
- Input Devices : Keyboard, Mouse.
- Ram : 8 GB.
SOFTWARE REQUIREMENTS:
- Operating system : Windows 10 Pro.
- Coding Language : Python 3.10.9.
- Web Framework : Flask.
- Frontend : HTML, CSS, JavaScript.
REFERENCE:
Prasanna Blessy; K. Kathiresan; N. Yuvaraj, “Deep Learning Approach to Offline Signature Forgery Prevention”, 2023 9th International Conference on Advanced Computing and Communication Systems (ICACCS), IEEE Conference, 2023.
