Featured Post

Welcome to the Forensic Multimedia Analysis blog (formerly the Forensic Photoshop blog). With the latest developments in the analysis of m...

Wednesday, October 7, 2009

Ikena review - part 3

One of the things that we do with Photoshop and our other tools is to try to make images look their very best so that we can compare unknown patterns and objects to known patterns and objects in other photos - known as "comparisons." In the third test, I took a video of a subject that was moving and isolated the logo on his shirt. The video was taken in less than ideal lighting conditions but has 30 fps.

Again, working down the Enhancement tab, I set the Matching Frames and the Super Resolution functions to maximum. The logo was a small part of the scene, maybe 60 pixels square.

The focus was a little off so I set the Deblurring level to the maximum and increased the Focus level. I started to get a little "haloing" around the contrast cut off points (where dark meets light in the image), so I backed off a bit. I didn't want to create a modern art master piece, I just wanted the logo as clean as possible.

Not noticing much of an improvement, I tried using the Contrast adjustment. Again, not much improvement.

Here's what I started with.

Here's where it ended up.

Not much of an improvement at all. Yes, it's bigger. Yes, the contrast has increased a little. No, I still can't tell what it is at this point. If I had a known logo to compare it to, then I might be in luck. There's enough distinction and points to reference in the original. The final version's just bigger and darker.

Without a known copy ...

Verdict: D+

There's nothing here that couldn't be done in Photoshop for a lot less $$$. When it comes to revealing fine details, clearly Ikena doesn't handle out of focus, moving video very well at all - in spite of claims to the contrary. All the super resolution in the world isn't going to bring out what was never recorded in the first place.

In the real world, you get what you are given. During stings and other surveillance operations - you get one chance at the shot. The operators do their job and hope that we can improve on whatever they bring back. In this case, there's not much of an improvement at all.

Stay tuned for the next installment ...

No comments: