Hollywood and image processing

Hollywood distorts everything, but especially likes to make up its own science. People writing the scripts for these movies clearly have no understanding of even basic science. They read too much stuff on the Internet about quantum mechanics, radiation, genetics, you name it. They then proceed to make up their own explanations for what they read, and give these explanations to their audience. The biggest propaganda machine the world has ever known is working hard to misinform the world about science.

Image analysis always gets a good make-over from Hollywood. I believe it all started with No Way Out. In this movie, Gene Hackman accidentally murders his mistress, and tries to frame Kevin Costner, her other lover, for it. Costner is in charge of the investigations, and needs to expose Hackman. The backing of a Polaroid photo was found in the crime scene, and under the assumption that it must be a photo of the murderer, it is being “analyzed” by the crime lab. They have state of the art computer stuff there, of course. All of this is mid 1980’s. There’s a big screen with the blurry photo, becoming less blurry as the movie progresses. Costner knows all along that it’s him in the photo, and he knows how much time he has before everybody else does too (several days, I seem to remember). Very exciting stuff. But there’s no rhyme nor reason to the whole photo thing: Why is the backing a blurry copy of the photo? How are they able to deconvolve it so well? Why does it take such a long time to deconvolve? Why is the image refreshed so slowly after every iteration? I’d love to have that algorithm, though. I’d be rich!

Similar things happen over and over again in these movies. The worst offender to date, in my opinion, is Enemy of the State. In this movie, people are able to take footage from a security camera in a store and change the point of view. They find the footage with some guy they’re looking for, and pan around the scene to look at him from the other side. I seem to remember they were even able to look inside his bag! You can say that all of this is just good fun, but it does affect people’s perception of what is possible and what is not. No matter how good you are at image analysis, your skills cannot match people’s expectations. We then have to disappoint people asking for help, and end up looking bad in the process. Luckily, there’s still some people that realize that what they see on CSI is usually not realistic.

5 Responses to “Hollywood and image processing”

  1. On November 17th, 2009, at 12:52, Filip Malmberg said:

    Wow, those CSI guys are amazing. So, remind me again, how do I do this in MATLAB? I already tried the enhanceImage() function, but it doesn’t seem to work. 🙂

  2. On November 17th, 2009, at 12:56, Cris Luengo said:

    That’s because you didn’t set the parameters right… 🙂

  3. On November 17th, 2009, at 13:01, Filip Malmberg said:

    Oh crap. Does this mean I have to read the manual?

  4. On January 19th, 2010, at 14:17, Robert S said:

    Oh dear God no, I am writing a blog post 🙂

    Dare I say it, one has to be a boob to actually believe “hollywood-science”

    is “real-science”. No offense.

    I have wondered. Do image analysis lecturers ever get employed by, lets say, the local police to advice in CSI like matters?

  5. On January 19th, 2010, at 17:49, Cris Luengo said:

    Robert, I think you over-estimate the audience of these programs! 🙂

    I once had to put a photo negative under a microscope to show a police investigator that there was no way of reading that license plate. It covered about 5 or 6 crystals.

Leave a Reply

Your email address will not be published. Required fields are marked *

Note: I moderate all comments. Comments without a clear relation to the text above will not be published.

You can use these HTML tags: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong> <pre lang="" line="" escaped="" cssfile="">