CCTV in Criminal Investigations

This info graphic has been produced by JMC Technologies and was posted today at IFSEC Global. I thought it was quite good but missing some key points.

how-cctv-is-used-in-criminal-investigations

In the ‘Common CCTV Problems’ section, there is one VERY big point missing…

COMPRESSION

This is often overlooked and can destroy any other previous attempts to gain an image of evidential quality. I regularly see footage that should offer substantive information but due to the compression, the details have been discarded.

Even when recording at higher than SD resolution, if the compression throws away the detail, any benefits are lost.

It’s good to see that BLUR is in there. Again, one of the common issues I have to attempt to restore is Motion Blur. You can’t just stick a camera up and hope for the best! If the operational requirement for the camera is to capture persons as they leave the store, or capture a vehicle as it exits a car park – remember that those objects will be moving! The camera has to be able to capture that fast moving object.

Lastly, and again it’s missing from the info, is that after the evidential data has been exported from the recording device in its native format, it must be immediately playable without the requirement to install proprietary codecs or players. Can you imagine how many hours are lost worldwide with officers and staff just trying to play CCTV?

It must be understandable and open to validation and investigation. I’m not talking here about a manufacturers own ‘authenticate image’ button. Forensic Video Analysis starts with the independent analysis of the raw data. I have learnt over the years that players do all sorts to the image, from distorting the aspect ratio, softening the image with a hidden deblock filter, and even skipping frames.

It’s not just Video – It’s Evidence, and should be treated as such!

2 comments on “CCTV in Criminal Investigations

  1. Compression is usually the biggest issue. A case I am curently working, the incident was was imaged by an SD and an HD camera and the SD has proven more useful due to the less aggressive compression applied. Really frustrating!

    I noticed a few other issues with the graphic. The figures quoted under the performance standards for CCTV section are based on screen size. Clearly the spatial resolution is more relevant than the screen size. The figures are also massively dependant on so many other factors (compression, lighting, the features of the persons imaged – think a tattoo on the neck) that they become arbritrary and meaningless.

    Also, under the guidelines it mentions ‘There must be no evidence of editing’. As you know, editing is often useful, if not necessary, to give a clearer, more concise overview of what occurred.

Leave a comment