Face Recognition
SmartVision is an advanced video surveillance system that includes a face recognition feature based on machine vision. When motion detection is enabled, the system automatically activates face analysis, allowing for the accurate identification of individuals in the frame. Face recognition enhances security by improving access control and providing an automatic response to key events, such as the appearance of specific individuals in the surveillance area. This feature works on 64-bit computers with sufficient computational power and can be used separately from other object detection methods.
Setting up Face Recognition in SmartVision
In the IP camera settings, select the "Detection" section. Here, you can enable both object detection and the face and license plate recognition module.
Enabling Face Recognition
If you have already activated motion detection on the previous tab, the face recognition feature is also available in SmartVision.
This feature allows the system not only to detect motion but also to identify faces of people appearing in the frame.
Face Recognition in SmartVision
To enable face recognition, you need to activate the "Enable face recognition" option.
This feature can work in conjunction with object detection, but it can also be used independently without enabling other object detection methods.
This feature allows the system not only to detect motion but also to identify faces of people appearing in the frame.
Face Recognition in SmartVision
To enable face recognition, you need to activate the "Enable face recognition" option.
This feature can work in conjunction with object detection, but it can also be used independently without enabling other object detection methods.
How Face Recognition Works:
Technical Requirements:
For proper operation of the face recognition feature, a powerful computer with sufficient computational resources is required. This option is only available on 64-bit systems, as the face recognition algorithms demand significant processing power.
Advantages of Using Face Recognition:
- Motion detector trigger – When motion is detected, the system activates the face recognition process.
- Face identification – The system analyzes the image and attempts to recognize a face in the frame using machine vision algorithms.
- Event recording – If a face is recognized, the corresponding event will be recorded and saved.
Technical Requirements:
For proper operation of the face recognition feature, a powerful computer with sufficient computational resources is required. This option is only available on 64-bit systems, as the face recognition algorithms demand significant processing power.
Advantages of Using Face Recognition:
- Increased security: The system can not only detect motion but also accurately identify who is in the frame, which is useful for access control and security.
- Tracking key events: Face recognition allows for quick and accurate retrieval of recordings, such as for analyzing visitors or employees.
- Automatic response: Depending on the settings, the system can automatically respond to recognized faces, such as sending notifications.
Face Recognition in the Events Form
In this section, you will see all motion detection triggers, as well as object detections and specific recognized faces.
You can also apply a filter to search for the desired events through the search form.
The events form also displays the status of file uploads to the cloud server if video transfer is configured for the selected camera.
Face recognition in SmartVision is a powerful tool to enhance security, providing additional capabilities for video analysis and access control management.
You can also apply a filter to search for the desired events through the search form.
The events form also displays the status of file uploads to the cloud server if video transfer is configured for the selected camera.
Face recognition in SmartVision is a powerful tool to enhance security, providing additional capabilities for video analysis and access control management.
Face Recognition Database
You can use photos to train the neural network and create a face database that will be used for real-time recognition.
The training process for a single face is very simple and takes only about 5 to 30 seconds.
All you need to do is upload several images of that face into the system, and the neural network will begin training to recognize it in video streams.
The training process for a single face is very simple and takes only about 5 to 30 seconds.
All you need to do is upload several images of that face into the system, and the neural network will begin training to recognize it in video streams.
Adding New Faces for Recognition
To add a new face to the database, start by entering the name of the person in the corresponding field and click the "Add" button. After that, select at least 5 photos of this person, which the system will use for training. Make sure the photos vary in angles and lighting so the neural network can better recognize the face under different conditions.
Once you have selected the photos, click "Save & Update" to save them and start the training process. The system will process the images and extract key facial features necessary for further recognition.
Once you have selected the photos, click "Save & Update" to save them and start the training process. The system will process the images and extract key facial features necessary for further recognition.
It’s important to note that the database does not store the actual photos, but only their "digital weights"—unique data that the neural network uses for precise face recognition. These digital representations enable quick and efficient face search and identification in video streams, allowing the system to recognize a face even if it changes due to angle, lighting, or other external factors. Wait for the process to complete in order to update the database and start using the new face for recognition.
Additionally, the system has the capability to automatically create a face database based on video streams from surveillance cameras. For this to work, the camera must capture at least 5 images of the same face from different angles. The system will analyze these images, extracting key features to efficiently recognize the face in the future, regardless of angle or lighting.
This approach allows for the quick creation of a face database without the need for constant manual image uploads, greatly simplifying the setup and training process for the recognition system.
Additionally, the system has the capability to automatically create a face database based on video streams from surveillance cameras. For this to work, the camera must capture at least 5 images of the same face from different angles. The system will analyze these images, extracting key features to efficiently recognize the face in the future, regardless of angle or lighting.
This approach allows for the quick creation of a face database without the need for constant manual image uploads, greatly simplifying the setup and training process for the recognition system.