Unique Visitor Counting🔗

The Unique Visitor Counting module is designed for building reports on counting unique visitors. It is possible to exclude from counting the persons present in the face database and included in the specified groups of this database. This may be required when you want to exclude your own employees from counting.

Capabilities of the module

  • Counting people wearing glasses, hats, etc.

  • Counting people in masks

  • The number of employees in the base is not limited

  • Determination of gender and age

  • Emotion Recognition

  • Reports

  • Using a high-performance graphics card (GPU)

  • Using multiple graphics cards (GPU)

Details

The Unique Visitor Counting module uses all video cards installed on the server and matching the requirements. In this case, each camera with the module running is matched to one of the video cards.

In a situation when the number of cameras with enabled recognition modules exceeds the number of video cards in use, cameras will be distributed to video cards evenly, without taking into account the characteristics of video streams from cameras and video card performance (i.e., an equal number of cameras will be attached to each video card, if possible).

The module uses the recognized face access control technology, in which the uniqueness of a face is determined based on a set of features called an index. In this case, the module does not operate on individual points of the face when determining the index; instead, the entire 100x100 pixel image is analyzed.

The Unique Visitor Counting module can identify masked faces of employees with a fairly high accuracy; provided that the database contains samples of unmasked faces of these employees. Also, this plugin can identify turned faces; even if the database contains only images of these faces looking directly into the camera.

The face database can be located either on the same server where the recognition is performed or on another server of a single multi-server video surveillance system.

Each record in the face database contains:

  • one or more images (screenshots and/or photos) of the person's face;

  • surname, first name, patronymic (full name), which is entered manually and is an optional parameter;

  • additional information (optional text field);

  • group membership (optional parameter);

  • the uniqueness of the records of the person database is determined by the person index. Thus, the database may contain several records with matching full name (including those with empty full name value).

You can work with the face database either from the Eocortex Client application or using the API.

Compatibility with other modules

Requires Eocortex motion detector

Neural Networks

Compatible with modules

Incompatible with modules

Standard

Special

√

√

√

  • Abandoned object detection module

  • Auto Zoom

  • Counting People in Queue

  • Crowd Monitoring

  • Emergency Vehicle Detection

  • Face detection

  • Face Mask Detection

  • Face Recognition (Complete)

  • Fire and Smoke Detection

  • Frame Area Blurring

  • Loud sound detection

  • People Counting

  • Personnel Activity Monitoring

  • Sabotage Detection

  • Search for Objects

  • Shelf Fullness Check

  • Traffic density heat map

  • Tracking

  • Uniform Detection

  • Face Recognition (Light)

  • Fisheye dewarping module

  • License Plate Recognition (Complete)

  • License Plate Recognition (Light)

  • Object Classification and Counting

√

supported and required for the module to work

+

supported and provides additional features of the module

–

not supported or not required for the module to work

⚠

not recommended for use with the current module
Usage

Using the module in the Eocortex Client application.