Face recognition function on Synology DVA3219 | Test and review

Many corporate customers of Synology NAS choose these solutions for video surveillance systems, because since time immemorial, this company has offered a simple and effective tool for ensuring security in the home, in the store or in the cottage. And since the company’s business is active, last year it offered a video surveillance solution with intelligent face recognition.

DVA3219 - the first device for Edge video surveillance

Now the IT industry is experiencing a real boom in Edge devices, and Synology DVA3219 is a typical brick for building smart video surveillance “somewhere out there”, on a remote object, whether it is a warehouse, store or office. It is not necessary to compare this model with other Synology analogs, this is a completely separate direction for the development of the company’s model range.

Synology DVA3219 specifications:

  • Intel Atom C3538 processor (4C, 2.1 GHz, Hardware AES-NI)
  • RAM: 1 4Gb DDR4 module, up to two SO-DIMMs with a total capacity of up to 32 GB
  • GPU: Nvidia GeForce 1050 Ti, 4 GB

Disk subsystem:

  • 4 bays for 3.5/2.5" HDD/SSD with SATA-600 interfaces
  • Expansion with two disk shelves with DX517, 5 HDDs/SSDs each
  • Support for SSD caching

Interfaces:

  • 4x1G Base-T 1000Mbps Ethernet (LACP, VLAN)
  • 3x USB 3.0

Integrated power supply

The platform itself for desktop NAS-s at Synology has not changed for a long time, so I do not see the point in describing the design of the case and cooling. Instead, let’s answer the most common questions about how this device works.

Can the DVA3219 work as a NAS?

Yes, this MODEL supports all the same features that are implemented in the DS series desktop NAS. You can use file sharing over CIFS (SMB) / NFS / iSCSI protocols, backup your IT infrastructure with Synology Active Backup for Business, virtualization platform, container virtualization, media server, and even download torrents, if this is important to you.

Which video card is installed in DVA3219?

The most common NVIDIA GeForce Ti 1050c 4 GB of GDDR5 memory from the OEM, which could not be determined. The video card can be pulled out and installed in a desktop PC: it is detected, you can work and play on it by connecting a monitor, if of course you can find something to play on the GeForce Ti1050.

But you can’t connect the monitor to the NAS: the HDMI and DVI ports are hidden inside the case.

Can I use the video card for something else?

when you first set up the NAS, you install the CUDA SDK in the system, but you don’t have a choice of version. When using Docker, it is quite possible to get access to GPU resources, but given that CUDA presents GPU resources completely for the application, intelligent video surveillance functions will not work.

Are the functions of deep video analysis available on models with powerful CPUs?

No, such features as face recognition, perimeter detector, and client flow counting require a different computing architecture, and can only be run on a NAS with a GPU.

What else does DVA3219 differ from the usual Synology desktop NAS?

Since the device is positioned as a peripheral video recorder, there are no dedicated slots for SSD cache and there are no high-speed 10-Gigabit interfaces. But the device has as many as 4 network ports, for each of which you can configure a VLAN to differentiate network traffic and channel aggregation to increase fault tolerance.

What are the smart features?

Of course, the flagship intellectual function is the ability of the device to distinguish people on video, so that you can make a file of your visitors and employees, and then see who and how often fell into the field of vision of the camera.

DVA3219 can run 4 video stream analysis tasks at the same time, whether it is 4 cameras or 1. These are, in fact:

  • face recognition,
  • a counter for the number of visitors,
  • detection of people and/or cars in the zone,
  • tresspassing through the virtual border,
  • more accurate detection of traffic (taking into account weather conditions)

Face detection function

Perhaps we should start with the fact that Synology has learned to distinguish people from animals and cars on the record. But in the current version of the software object belonging to a race of people, machines or animals, the device determines solely on the physical dimensions in the frame: if movement is recorded on a small area of the frame, in the database the event is recorded indicating the animal on the middle of the square people in a great car. Therefore, if an elephant or a horse gets into the frame, there are possible options… But this, as they say, is only a pleasant digression, but in fact, at first you will have to spend a certain learning process, creating a database of portraits of your visitors.

Technically, tests have shown that the algorithm focuses on a person’s eyes or glasses when recognizing a person. Therefore, if you hide your eyes from the camera, the facial recognition algorithm will not even attempt to recognize you by secondary features, such as cheekbones or the shape of your nose. In addition, the color of a person’s clothing, or rather its contrast with the surrounding environment, is of great importance. If your clothes match the tone of the walls, and may not recognize you (which is not so bad). But if the NAS recognizes an object in the frame, it continues to “guide” it, and even if the face suddenly becomes indistinct (for example, due to dirt or insects hitting the lens), the object still remains registered, which is clearly visible in the following screenshot.

You ask me how to apply face recognition locally? It is quite simple: you can create a database of your employees and guests, determine their status (for example, VIP or Vice versa persona non grata), and then easily select the employee you are interested in in the database and open all the records in which he, frankly, lit up.

Also, a separate column records unknown people, which may be ordinary couriers or rare visitors. At first, there will be cases in your database when the system does not recognize your registered users, for example, due to poor lighting. We recommend that you periodically review such situations and manually add an illegible image to the user’s account, so that the next time the system will be easier to identify the person under unfavorable shooting conditions.

And do not think that face recognition requires some kind of ultra-expensive 4K camera: the resolution of 720p @ 24 FPS is enough for this function to work confidently. In principle, Synology DVA3219 already has enough features to open doors for employees automatically based on facial recognition.

Additional features in-depth analysis

DVA3219 can run 4 video stream analysis tasks at the same time, whether it is 4 cameras or 1. This is, in fact, face recognition, analysis of the number of visitors who have passed through the door, video search for cars or animals, and analysis of violations of the virtual border.

Of course, our imagination, fueled by rumors about the upcoming digital monitoring of the population through surveillance cameras, draws truly limitless possibilities for this technology: we dream of searching for a person by photo, cataloging his movements around our object where video surveillance is conducted, and the possibility of searching for his profile in social networks. Yes, but… the reality is much more prosaic. Today, the Synology DVA3219 functionality allows you to detect people on video and… view the latest events by events in the shared feed.

This feature can be especially useful when you are observing a large area, such as the courtyard of a building, and in the General flow of events, it will be easy to determine when people came into view of the camera, and when - vehicles. Later, when investigating incidents, this can reduce the time spent on finding videos.

In fact, the ability to separate people from cars gives rise to two additional functions of intelligent video surveillance: counting people crossing a virtual border, and improved motion detection. Today, working with human flows, the ability to predict the direction of the crowd in different conditions, is almost the basic pillar of urbanism and advertising. But in order to be able to count the number of visitors passing through the door, the camera must be installed strictly on the ceiling and look at the tops of the heads. We don’t have this option, so we can’t test it. By the way, today video surveillance technologies allow you to count the number of visitors and build heat maps of site attendance, so there are no technical restrictions for this function, and Synology needs to be corrected urgently.

What should I do about it?

Some readers naturally had a question: well, Synology recognizes faces, and what to do about it? But in fact, a very correct question. Synology video surveillance systems provide the ability to create scripts describing the event logic. For example, “motion detected in the camera…” and then there are possible options. For example, from the built - in templates, you can make a sound signal on the video surveillance server, take a screenshot, or even open the door, but this is only part of it. Much more interesting is the ability of the server to “knock” on a web address with a GET or POST request, which will give you the simplest option for authorization or enabling a device.

But the most important thing is the support of the IFFTTT platform, which opens the door to the vast world of the Internet of things (IoT), and here you can program the events of your electronic assistants in two or three clicks, depending on the Synology signal. Well, starting from a banality like “turn on the kettle if the boss passed through the checkpoint” and ending with various notifications via SMS, E-Mail, Push, opening/closing smart doors, locks, managing “dry contacts” of expensive IP cameras, and so on. IFTTT is a popular platform for simple scripting algorithms that is supported by thousands of different devices.

Of course, Synology did not flaunt such a tidbit of artificial intelligence, but shoved it into the far corner of the Surveillance Station menu: click the “start” button in the upper-left corner and select “action rules” with an icon that somehow resembles a calendar. Next, we create an event that will be generated by face recognition. Here is the full set: whether a VIP object appeared in the frame, whether an unknown person or a blocked user appeared. On the next tab, we select an action, and that’s it, the setup is complete.

Among the built-in profiles, there is a very important “medical mask detector” that will help you raise discipline during epidemics. The only thing that is not enough is the built-in profiles: I would like to see ad placemakers, people in uniform, smoking, not wearing a mask or helmet, and it is better to give the opportunity to create such conditions yourself. But, this is not even in expensive commercial software, what can you expect from a NAS?

The prospect of integration in Surveillance CMS

Of course, you can only imagine what prospects are provided by using devices such as DVA3219 as dedicated servers for AI image processing in distributed video surveillance networks. I really hope that in the future Synology will allow NAS to recognize faces on recordings made by completely different NAS-es in other branches in other cities. And in General, for such purposes, it will be possible to centrally install NAS-s with the GPU and conduct remote analysis of Edge video, even with a small delay. All this promises great prospects in the future.

Well, today the only way to use the DVA3219 AI functions in a distributed CMS network is to install this NAS as the main host server, because this model can use intelligent video analysis functions only for recordings from its cameras. If you connect the device as a video recording host, you will need to log in to the server itself to access the smart features.

Conclusions

Today, DVA3219 is a ready-made solution that can improve the quality of your enterprise security service, while also working as a backup storage, and I have no doubt that Synology will add GPU installation to its enterprise solutions, which usually have empty PCI Express expansion slots. Today, this is a viable solution that can increase the security of your company and make it easier to establish the circumstances of the incidents that occurred.

Practically, now you can track the movements of your employees, get information about visits to your retail outlet, or record the passage of automotive equipment in the database. This is a very good start for a desktop NAS, which still retains all the functionality that users buy Synology devices for.