SSIM: An Automated Solution for Signature Verification in Digital Systems
Abhilash S Nath, Senthil Pandi S, S. Venkata Lakshmi, L Preethimathi · 2025
This research presents an auto-approach to verifying signatures on documents using image processing techniques. Structural Similarity Index (SSIM) is employed by the system to measure the closeness of two signature images. Here, we convert the images to grayscale (so we’re only considering structural information), and fix them all to the same size (for ease of comparison), then calculate the structural similarity between the images and get a similarity percentage back that tells us if the signatures match or not. This method provides the simplicity and effectiveness that allows for the recognition and verification without having to be complicated with the complexity of machine learning models. We significantly reduce the time needed for manual verification with this solution, while maintaining high reliability. Due to this, it is ideally suited to practical use cases where both security and speed are necessary. This system uses SSIM to make it robust, allowing errors in slight editions of the signatures of genuine and forged signatures to be detected with high accuracy. Further, it decreases the human error rate during the verification process.