Have you experienced annoying “stops” while watching YouTube or Netflix?

People normally blame the network for these few seconds of delay. It is indeed due to lack of resources in the delivery infrastructure, which can be caused by the network, server or storage.

What does this have to do with surveillance?

With all IP cameras, the lack of resources happens in the video surveillance infrastructure too. It occurs as often, if not more often. When these “stops” appear, we see frame drops or recording gaps. In other words, the infrastructure can degrade the video quality. If the infrastructure is leaky, or too many cameras’ traffic going through the networks, the video quality at the end can be very poor.

Is this a problem?

Most people don’t know that they are missing video until something bad happens and they try to access the recordings. If a crime occurred, key images may not be there. This can impact the ability to defend in lawsuits and can result in the loss of properties or lives.

What is the video quality specified today?

We primarily see the video quality specified at the source, where the cameras are, measuring things like HD resolutions, the frame rate per second, the choice of compression, etc.

What should we do?

The video quality has to be end-to-end. The video that we see or store at the end-point is what really matters. The typical end-user doesn’t have this knowledge and there are no industry end-to-end quality guidelines. People need something simple to ask for.

How about using AI/Robot to watch video?

Artificial intelligence, like facial recognition, demands high video quality to reach its full potential. For every camera watched by AI, we need to be ready to provide quality video.

What can Rasilient help?

Over the years, Rasilient has been working hard to ensure No-Frame-Drop. Please check out our NFDMeter, which reports frame drops at the recording end.