How to organize the work of cameras in a warehouse?

Question

CCTV

There is a mobile application used by order pickers in the warehouse. It is also planned to conduct video surveillance at the warehouse and record the assembly process of each order.
Those. the scenario is as follows - the order picker logs in to the application under his credentials.
Enters the number of the order to be collected.
Video cameras that are responsible for certain types of goods are turned on (depending on the order number, it is always clear what goods are in it and the system knows which cameras need to be turned on).
Cameras record and attach them to this order.
As a result, the Administrator can view the entire assembly process from each camera through the web interface in the order history. Those. the video must be stored on the server and it must be possible to view it.
The question in architecture is how to optimally organize such a process?
Is it possible to manage video surveillance through a CRM web server that is physically located far from the warehouse (maybe the cameras need to be bought with an autonomous wifi connection and a remote control interface or something else?) Or is it necessary to install a separate server that stores the video information and from which it is received by the already remote server?
Thanks in advance for your advice, thanks.

Answer 1

and several questions immediately arise:
1. Start of recording - at the start of the formation of the order. And the finish? Is there any trigger?
2. Where are you going to store the video?
Cameras themselves do not traditionally store video on their own. They broadcast it to the server.
Those. in your case, the server itself (CRM or video surveillance, more logical - the second) decides to save the video - and starts / finishes recording the stream that the cameras give it.

Answer 2

In the topic of video surveillance, I don’t rub, but I will express some of my thoughts.
There is a magazine "Security Systems" or similar in the field of video surveillance. It makes sense to read several issues for a better understanding of the topic.
The question asked has a very mediocre relation to web development.
It is limited to the interaction of a web server with web clients, and all the nuances of the network architecture should fall on the shoulders of network engineers with experience in video surveillance.
You can manage some physical devices through the web interface, but in the end it will all come down to building a layer for working with peripherals and building an interface around the layer.
Are the cameras actually controlled from the outside or with a fixed viewing angle? Is there anything to manage?
And what about CRM?
IP surveillance cameras should be located near the IP video server, which should be located in a closed, cooled and dust-free room on the same floor. With Wi-Fi, there are big doubts about the quality of communication, especially since many other radio sources can work in the warehouse, causing interference, as well as potential problems with the electrical network. The cable is more reliable if properly laid and protected from external influences. You can choose Cat6 cable to connect the cameras to the DVR, and Cat7 cable to connect the DVR to the surveillance point. Or even fiber.
Since the web server acts as a command and control server, it can be placed separately from the DVR.
In general, it's still better to read
www.cyberlink.com/support/product-faq-content.do?i...
habrahabr.ru/post/177071
Search keywords: DMS (Digital Media Server), RTSP, IP camera.

Answer 3

You need a video server with API support for integration with a mobile application. Conventional video servers and some DVRs can be event-driven, such as triggering recording when motion occurs in a fixed area. Where the video server will be physically located is not so important, the main thing is to provide the required network bandwidth. Wi-Fi is not very suitable for organizing video surveillance.

Answer 4

IP cameras, rtsp stream. At the command of the software from the side of the assembler, "build started" - we rob the stream into an ffmpeg or vlc file (though stupidly shell command), at the command "build finished" - we complete the recording and load the file / link into the database. If the pickers are not forty people, everything will take off perfectly. Video servers, refrigerated rooms, controlled cameras should be left to the security guards.
Or, alternatively, if the SVN already exists, you can request a piece of recording from the video server and load it into the database using the known time frame and the assembler login (login->workplace->camera numbers) using the command to complete the assembly of the order.