Requirements and recommendations for the Face Mask Detection module

Detection

The module is capable of detecting up to ten people not wearing face masks in the frame at the same time (if allowed by the computing capacity).

The module does not recognize (identify) faces, it cannot tell one person from the other or compare a person’s face with the faces from a database; it only finds people without face masks in the frame.

When an infringement event (no mask) is detected, the module briefly highlights the person’s face with a red square in the client application and creates a corresponding event in the event log. The repeated detection of the infringement by the same person will become possible only after the disappearance of this person from the frame for three seconds minimum (e.g. when the person leaves the frame or covers his/her face completely).

Hardware and software

Warning

The Eocortex Neural Networks package must be installed before it will be possible to use neural networks-based features of the module.

The following equipment is required to use this neural network-based module:

• A processor that supports AVX instructions;

• An NVIDIA video card (GPU) with the computation capacity index of at least 6.5 and with at least 4 Gb of memory; the parameters and performance of the video card must be similar or better than those of NVIDIA GTX 1650 Super model;

• Version of the video card driver at least 460;

• Swap file at least half of the total RAM size.

If the package will be installed on a virtual machine, it may additionally be required to:

• Enable support for AVX instructions in the guest machine settings;

• Use GRID drivers for GPU virtualization.

Warning

Eocortex must use video cards selected for running neural networks in monopoly mode. It is not allowed to use such card for other applications or tasks that consume GPU resources, including for displaying video. Simultaneous use of a video card for several tasks may lead to incorrect system operation: from analytics performance degradation to server instability.

Warning

The neural network works with the 64-bit version of Eocortex only.

Warning

When upgrading Eocortex to another version, it is necessary to also upgrade the Eocortex Neural Networks package to the corresponding version.

Module settings

Setting too small minimum face size may lead to false positives due to low image quality.

The shape of the detection area can be arbitrary.

Masks

The examples of supported face masks:

The supported colors are as follows: white, black, yellow, different tones of green, blue, and red.

The face covered by a hand, a scarf or a mask will trigger the detector if the nose area is not covered. However, if the nose area is covered, it will not trigger.

Video stream

• Image resolution: no lower than HD (1280x720).

• Framerate: 15 fps or more.

Lighting and image quality

Illumination of faces in the frame must be uniform and constant.

If the camera is installed opposite a bright source of light (sun behind the entrance door, etc.), it is required to adjust the exposure or brightness in such a way that the face in the frame is light. The overexposed background is acceptable.

The image quality must be medium or better. Significant compression artefacts are inacceptable.

No blurring of moving people’s faces is allowed.

The image must be in color.

Scene and camera position

Faces must be fully and clearly visible in the frame.

Persistence of reflective surfaces (glass, mirror, other objects giving reflections) in the frame is not allowed.

Strong lateral illumination (e.g. sunlight from the window) resulting in the overexposure of one part of the face is not acceptable.

The camera may be placed above the face level, directly facing the people to be recognized. In such a case, the camera elevation angle must not exceed 35°.

The distance between the pupils of a face to be recognized must be at least 30 pixels.

The camera must directly face the people to be recognized. The camera angle between the face direction and the lens axis must not exceed 30°.

Examples of non-compliance

Covering (overlapping):

✓	X

Face illumination:

✓	X	X

Blurring of face in motion:

✓	X

Unacceptable distance between pupils:

30 pixels	15 pixels
✓	X

Video stream quality:

✓	X

Camera tilt:

35°	50°	70°
✓	!	X

Lateral light source:

✓	X

Reflecting surfaces:

✓	X

Horizontal turn of camera from face plane:

0°	15°	30°	45°	60°	90°
✓	✓	✓	!	X	X

Black and white image (night mode):

✓	X

Recommended camera settings

Quality: highest.

Bitrate: maximum possible.

Profile: maximum possible.

I-frame interval (GOV): 50.

Stream anti-aliasing: off.

Exposure and brightness: ensuring that the face is clearly seen (if the camera faces the light source, the overexposed background is acceptable).

Shutter speed: must not be too low (more than 1/50), because in such a case the blurring of moving objects will occur.