I’ve posted many articles about intelligent video.  There’s a lot to it, there are many nuances, and there are many options.  In an effort to start tying the whole picture together this post is a brief explanation of the best practices to observe when implementing intelligent video.

In the near future I will also write about the best practices for using intelligent video applications like license plate recognition or facial recognition.

Here are the 5 best practices for understanding the framework behind a successful intelligent video solution.

5 Best Practices

System Reliability

The system framework should minimize risk of failure and downtime.

Scalable and Flexible

The system should be scalable and flexible enough to scale up from a few cameras to many cameras and it should be intelligent enough to divy up processing loads on different network components.

Interoperability

Is the video management software and the intelligent video application you want run limited to only certain hardware that you’re using or, can you install equipment from different vendors? 

You should know your options, or limitations as it were, beforehand so you that you’re not stuck with unrelistic expectations.

Data Format

Understand if the intelligent video metadata is based on open source or not before trying to incorporate it into your other business systems.

Accuracy

How accurate is your intelligent video?  Well, keep in mind that most intelligent video systems are NOT 100% accurate and your surveillance personnel should be prepared for false-positives and too many will become overwhelming.