CASE STUDY

Description

An expandable solution, for time registration, based on face recognition. The device performs automatic identification of a person and has the ability to generate employment reports. There is possible integration with third-party extensions and binding with running applications on PC. A valuable feature of the app is that it scans launched applications on the desktop, binds them with individuals, and generates reports.

Challenges

The team faced the task of implementing face recognition with limited hardware performance.

Another challenge was to pick up an appropriate cluster to implement a face recognition project.

Recognition speed and overall performance were important in order to accurately calculate time without accumulating errors.

Responsiveness and resistance to high loads.

Multiple face recognition support.

An important task was to determine the currently launched applications and, accordingly, to log the time that the user spends in an application.

Approach

Embrox decided to split the project into separate sub-projects, among them: back-end, front-end, the desktop listener.

The Team has created a platform, which works on desktop PC (with the usage of regular USB camera) as well as on limited Linux platforms (for instance, raspberry PI). Also, the system was designed to work asynchronously, so it computes captured images in the background.

Key features

• Registration of a new user (user management)
• User confirmation / merging recognition results
• WiFi access to the dashboard
• Report generation
• Possibility to extend desktop applications

Results

The application has started to be used in alpha testing mode.

This solution demonstrated that it is stable and can be used as an automatic logging system.