Vision Mobile and Web SDK

TRUST ID is a client-side server system designed for facial recognition and various types of photo analysis.

System functions:

• Detection of a face in a photo or its absence: The system uses a face recognition algorithm (Face Detection) to detect and highlight faces in the provided photographs;
• Comparing photos with data from the database: Using the 1 to N comparison algorithm, the system compares recognized faces with data stored in the database and determines the percentage of matches with each profile;
• Liveness detection (Определение живого лица): Для предотвращения мошенничества система использует алгоритмы для определения, является ли лицо на фотографии реальным человеком или статичным изображением, маской, видео или дипфейком;

Photo quality analysis:

• Photo blur: The system can analyze image blur, which allows you to evaluate the quality of the photo and its suitability for identification;
• Lighting and photo color: Lighting and color analysis algorithms allow you to take into account possible distortions due to shooting conditions;
• Eyes open/closed in the photo: The system can determine whether the eyes on a face are open or closed, which is also important for confirming a living person;
• Determining the angle of rotation of a face in a photo: The system can analyze the angle of rotation of a face in a photo, which helps in correct identification;
• Counting the number of people in the frame: Using object and face detection algorithms, the system can count the number of people in a photo;
• Probability of a photo belonging to a specific person: The system can estimate the probability that a photo belongs to a specific person based on comparison with records in the database;
• Custom settings: The user can configure photo comparison thresholds, acceptable blur, light, color and other parameters depending on their needs.

Algorithms used:

• Face detection: Face detection algorithm is used to detect the face in a photo;
• 1 to N (1 to N) comparison: The algorithm allows you to compare recognized faces with a database to identify matches;
• Liveness detection: Liveness detection algorithms help prevent fraud;
• Comparison: The system can use comparison algorithms to evaluate the similarity of photos and compare the data with the database.

This system provides a high degree of accuracy and reliability in identifying and analyzing faces in photographs, making it a useful solution for a variety of applications, including customer and employee authentication, and security in a variety of applications.