How to design a CCTV system that give ID quality results

[easyazon-image-link asin=”B000V5T4BU” alt=”Sony EVI D70 – CCTV camera – PTZ – color – 1/4″ – optical zoom: 18 x” src=”” align=”right” width=”75″ height=”75″]It is unfortunate that many CCTV surveillance systems do not provide surveillance footage that is of much use when it comes to identifying people. While the quality is improving with High Definition cameras, the fundamental errors are still common.

A frequent mistake is trying to get the surveillance camera to cover an area that is too wide, resulting in a lack of resolution on peoples faces. The images we see on TV shows like CSI lead us to believe any image can be enhanced but the reality is very different.

Here are some very simple to implement tips that will improve the likelihood of you getting the perfect mug shot of the suspect.

  • Have a camera at a choke point for ID and a second elsewhere for the overview
    This means as a person enters an area, they need to pass through a narrow area such as a door or corridor. This is the best spot to get a good quality face image. Once you have this, your other cameras just need to provide an overview of their movements.
  • Each camera should do only one job.
    A camera can either be used for an overview OR recognising a persons face. It is vary rare to be able to achieve both and get a good result
  • Think about what you would see if you were at the cameras height
    If you have a 3 mtr high roof, installing the camera up there will typically mean you get a great image of the persons bald spot or the brand name on their baseball cap. If you want to see their face, you will need to lower the camera. An angle of less than 20 deg is ideal.

There are many other aspects that can improve the image quality such as the camera sensor quality, compression algorithm, recording resolution and lighting but each is a subject of it’s own. To be notified of new site updates, join the mailing list on the home page.

Related Topics


Leave a Reply