Trust? Can you really trust an image?

Thursday, 28 September, 2017

Trust? Can you really trust an image?

By David Spreadborough, Forensic Video Expert, Amped Software

It will only be a matter of time before images like this appear due to Hurricane Irma, but a few weeks back, two images featured prominently in the initial reporting of Hurricane Harvey. The first was of a shark swimming along the Houston freeway. The second showed several aeroplanes virtually underwater at what was claimed to be Houston airport. These iconic images were circulated widely on Twitter and were featured on mainstream national media such as Fox News. There was just one small problem. Neither of them were real!

This situation prompts an important question. If this behaviour is widespread on social and traditional media then how do we know it isn’t also affecting police and court investigations? After all, if members of the public are prepared to manipulate images for the sake of a few likes and retweets, what will they be prepared to resort to when the stakes are much higher?

The well-worn phrases like, ‘seeing is believing’ and, ‘the camera never lies’ are deeply ingrained in our psyche, showing just how conditioned we are to trust a picture put in front of us. We very rarely only have an image. We are always told, or read, a context to that image. Our brains then begin to fool us into believing the image in front of us. However, this mentality now must be challenged. Every one of us has the means to alter an image on our computer or smartphone.

Image manipulation is, of course, nothing new. It’s been happening since the dawn of photography. There is also nothing wrong with manipulating an image. In CCTV and video enhancement, I do it all the time. The difference though is that I show a repeatable and reproducible workflow from the originals to my final image.

What happens then when we have no workflow, we have no documented list of changes? We have a single image and are simply asked to believe that it is a true and accurate depiction of that moment in time.

In my years serving as a police officer I would never have simply believed a suspect because they claimed to be somewhere else at the time of the incident. Instead I would have checked the alibi and sought other ways to verify this information. Yet digital images submitted as evidence are currently not required to go through any routine process of authentication.

On an individual case basis, this is already a problem. Depending on the incident, the importance of the image and the person conducting the investigation, the decision to have the image authenticated may never actually happen. Therefore, the image would be perhaps copied, printed and then maybe even photocopied into a court bundle and used as evidence.

As more and more of the evidence gathered originates from the smartphones of witnesses it is becoming increasingly difficult for agencies to receive and manage this visual data. For example, in a drive to cut costs, West Midlands Police Service is currently trialling an online self-service tool for crime reporting that allows digital evidence to be submitted.

From a CCTV and video evidence point, this brings one set of challenges in relation to the fact that many owners do not understand the importance of original video when it relates to evidence. From an image stance though, it brings an entirely other set relating to authentication. I am a huge believer in using technology to aid the processing of data through the evidential chain so I ’m not against this in principle. We must ensure though that we don’t cut corners when it comes to ensuring the chain of custody and verifying authenticity. After all, we must be able to trust what we see before us and what we finally present to a court.

As investigators it is imperative we start asking the right questions at the outset to ensure falsified images don’t waste valuable police time. Just like any other form of good police work this comes down to asking the right questions and making an informed judgment on the most efficient means of progressing the investigation. For example, who provided the image? Does it support or contradict those I’ve received from elsewhere? And is there any reason why they would benefit by offering up an alternative version of the truth for their own purposes?

I admit, image authentication was once a bit of a minefield. Various pieces of software, linked with the various scenarios. It is no wonder that the task of authentication has been shut inside the ‘Too Hard To Do’ Box!

Well, you can open that box now!

Taking all the tasks, tools, scenarios and filters required to analyse an image for authenticity, Amped Software have wrapped them all up into a single application developed to answer the needs of the investigative community. Amped Authenticate allows single and bulk actions to quickly determine which images have been supplied direct from a camera, and which have undergone some form of editing.

This, by itself, starts to save time and money but it is the full collection of tools, functions and filters though that truly make this a unique application for image authentication.

From bulk sorting to group data analysis. Selecting individual images to review and batch process. Validating meta-data and performing Camera to Image matching. Any function for image authentication, in one place.

When I get asked whether an image is fake, it is usually because the requester needs to trust it. They are asking me to conduct every possible test to ensure that trust. I can never say that it is authenticate, but I can state that after conducting nearly 80 different tests, I cannot identify any signs of manipulation or anomalies that would suggest tampering.

Do you deal with images from the public; do you Authenticate before using that image?

Perhaps it’s something that should be considered.

Return to news menu