AI with a Generic CCTV Camera

CCTV cameras are ubiquitous. You can find one everywhere. On the roads, at traffic junctions, in schools, malls, common residential areas, offices etc. Most of the CCTV cameras are dumb. They record what they see. Some of them save data or stream to a remote location. In this series of blogs I will show how you can make a generic CCTV camera intelligent with some Edge AI.

Ubiquity of CCTV cameras

As I had touched upon earlier, CCTV cameras are everywhere. As CCTV cameras get cheaper and people get more security conscious (or paranoid), we start putting them everywhere. The proliferation of CCTV cameras all around will have profound impact on our lives. We have to be aware that spaces that we considered private are not private anymore. The data has been recorded, possibly streamed already. It can be brought up at a later date for viewing or maybe viewed by a stranger remotely. This realisation is effecting the way we behave, in private and among others in subtle ways. I remember when CCTVs were first installed in our lifts, amorous moments of couples were often viewed by our security folks. Such incidents were discussed in a few community forums and I don’t hear of them anymore.

The proliferation of CCTV cameras will have the greatest impact on reducing crime. As more people get aware of the digital evidence we leave behind, criminals will realise that they can get caught and due to the evidence get punished more quickly than ever before. World over homicide rates – an indicator of violence and crime in any society is reducing. The Better Angels of Our Nature is a brilliant book that explored this in detail. Over the last few centuries societies methods of managing its deviants is getting more effective and better.

Smartening up your CCTV

I will stop digressing discussing the societal aspects of technology and specifically CCTV proliferation. We will stick to what we intended to do. Explore how AI can make cameras faster. In the coming few blogs I will show how to work with generic CCTV cameras and how they can made intelligent by adding some local compute to it. I will show how you can detect objects, vehicles, recognise people. I will demonstrate of how you can do this with computers and also with cheaper compute devices like raspberry pi and new hardware like Google Coral, Movidius Neural sticks, Jetsons nanos etc.

Signing till we see again..


Praveen Pavithran

Recent Posts

How to setup FTP on AWS Ubuntu

In this blog we will show how to setup an FTP on AWS machine. They…

1 year ago

How to run YOLO on a CCTV live feed

In this blog we explore how to run a very popular computer vision algorithm YOLO…

1 year ago

How to setup CI/CD for React using Jenkins and Docker on AWS S3

Introduction : Continuous Integration (CI) is a development practice that requires developers to integrate code…

1 year ago

How to setup a CCTV camera with JioFi

In this blog we explain how to enable live viewing for CCTV camera with a…

1 year ago

How to setup CCTV Camera with XMeye tool set

In this blog we will setup a Generic CCTV camera supported by XMeye. I will…

1 year ago

What is Edge AI and how is it done?

As explained in earlier blog edge AI is running your AI inference as close to…

2 years ago