| Age category (group) |
A group of people that is within a specified range of ages. In accordance with the age periodization of the World Health Organization, groups 18–44 (young adults), 45–60 (middle-aged adults), 61–75 (older adults) are distinguished |
| Attributes |
Age, gender, automatically determined by the system |
| Authorization |
A security mechanism to determine access levels or user privileges related to system resources |
| Avatar |
A visual representation of the face that can be used in the user interface |
| Best shot |
A frame of the video stream, in which the face is captured in the optimal angle for further use in the face recognition system |
| Biometric sample (sample) |
Analog or digital representation of biometric characteristics prior to biometric feature extraction and descriptor generation |
| Descriptor |
A binary data set prepared by the system based on analyzed characteristic. It is a composite vector of person's face attributes |
| Candidate |
Applicant for similarity with the reference |
| Cross-matching |
Many-to-many comparison (M:N). In the context of this document, comparing multiple lists |
| Task |
A task that is created by a user and runs in the background. In LUNA POINT, tasks include cross-matching, export, batch processing, batch import, and batch identification |
| Detection |
FaceStream entity that contains the coordinates of face and the estimated value of the object that determines the best shot |
| Event |
Detection recorded by the system with the extraction of attributes by the handler |
| Event card |
An information block displaying data about the event |
| Exchangeable image file format (EXIF) |
A standard for embedding technical metadata in image files that many camera manufacturers use and many image-processing programs support |
| Extraction |
A descriptor extraction procedure |
| Face card |
An Information block displaying data about the face |
| Face recognition |
A set of methods for collecting, processing, and storing data of person's face images for identity recognition or identity confirmation using mathematical methods |
| Faces |
Changeable LUNA PLATFORM objects that contain information about one person |
| Handler |
Image processing entry points that characterize the image processing procedure and define the LUNA PLATFORM algorithms used for this |
| Handling policy |
A set of rules (policies) for image processing |
| Identification |
Search for the most suitable descriptor by comparing the vectors of face features with a list of similar descriptors in the database (one to many) |
| List |
A set of faces in the LUNA PLATFORM system, combined automatically or manually according to a certain criterion |
| Liveness |
A software method to confirm the vitality of a person by one or several images in order to prevent spoofing attacks |
| LUNA PLATFORM |
Automated facial recognition system designed to process, collect, analyze, store, and compare biometric data obtained from facial images |
| Matching |
A procedure of matching descriptors for the purpose of comparison |
| Physical access control system (PACS) |
A set of hardware and software tools aimed at controlling the entrance and exit in order to ensure safety and regulate visits to a particular facility |
| Reference |
Object (attribute, face, face and event external IDs, event track ID, descriptor) that is compared/verified with the candidate |
| Similarity |
Probability characteristic in the range from 0 to 1, characterizing the level of similarity of subjects of biometric data |
| Spoofing attack |
Substitution of a real person for a fake image (for example, a photograph) to deceive the system |
| Track |
Information about object’s position (for example, face of a person) in a sequence of frames. A track represents a long continuous observation in which several events can be generated at certain intervals. It is needed to organize and aggregate data about groups of people who can appear and move in the video |
| Video analytics |
A set of functions that processes video frame by frame and evaluates useful data. Video analytics tasks can include people counting as well as detection of fight, weapon, and fire |