In this work, we present the development and analysis of a feature model for dynamic handwritten signature recognition to improve its effectiveness. The feature model is based on the extraction of both global features (signature length, average angle between signature vectors, range of dynamic characteristics, proportionality coefficient, average input speed) and local features (pen coordinates, pressure, azimuth, and tilt angle). We utilized the method of potentials to generate a signature template that accounts for variations in writing style. Experimental evaluation was conducted using the MCYT_Signature_100 signature database, which contains 2500 genuine and 2500 forged samples. We determined optimal compactness values for each feature, enabling us to accommodate signature writing variability and enhance recognition accuracy. The obtained results confirm the effectiveness of the proposed feature model and its potential for biometric authentication systems, presenting practical interest for information security specialists.

Keywords: dynamic handwritten signature, signature recognition, biometric authentication, feature model, potential method, MCYT_Signature_100, FRR, FAR

In this paper, we present the implementation of a neural network approach to solving the problem of handwritten signature recognition. We analyzed the main approaches to handwritten signature recognition. We identified the features of using a handwritten signature as an identification method, including the variability of a handwritten signature and the possibility of forgery. We identified the relevance of using neural networks to solve the signature recognition problem. We developed a neural network model for recognizing handwritten signatures, presented its architecture containing convolutional and fully connected layers, and trained the neural network model based on handwritten signatures "Handwritten Signatures" containing 2263 signature samples. The accuracy of the developed model was 92% on the test sample. We developed a web application "Recognition of a static handwritten signature" based on the developed neural network model on the Amvera cloud hosting. The web application allows identifying users based on a handwritten signature sample.

Keywords: handwritten signature, neural networks, signature recognition, image processing, machine learning, web application, cloud hosting, identification, verification, artificial intelligence

In this paper, we present a study dedicated to implementing a neural network approach to face recognition. We conducted a comprehensive review of existing face recognition methods. We developed a neural network model, trained on the DigiFace-1M dataset. This paper details the architecture of our developed neural network model and the step-by-step training process. The model achieved an accuracy of 78% on the validation set and 92% on the training set. We also addressed the integration of our model into the Russian Amvera Cloud service. As a result, we created a web application that allows users to identify themselves using uploaded images of their faces. This research demonstrates the potential of neural networks for face recognition tasks and offers a practical solution for implementing such systems in various fields.

Keywords: face recognition, deep learning, neural networks, user identification, model architecture, model training, model integration, cloud services, security, biometric technologies