High speed video in the science of "extreme events".

[drgjs]

Extreme events such as impacts, ballistics, explosions and blasts, which occur at very high pressures, and very high speeds, is difficult because of the highly specialized diagnotics, really involved physics and the computational methods which are generally very niche.

 

The one major advance in recent years is the advent and ever increasing availability of very high frame rate and high spatial resolution cameras in combination with the ability to store and process rapidly the very large amount of data generated, together with the development of batch image processing techniques as well as free software such as imagemagick, gimp, ifranview and many, many others.

 

This allows us to auto-analyze the data which is in the form of video - a sequence of images, run it through many different "filters" and "algorithms" rapidly to elucidate different features which are not apparent to the human eye. Yet, there are still many applications for which these readily available capabilities have not been fully exploited - which means there is all to play for and still much to learn from applying these technologies in areas of extreme science.

 

But as the digital camera technologies are still developing at a pace, as is computational science and power, including graphics etc, I do belieive that science in this area needs to constantly evolve with the times - we need to look again and again at what we might have videod ten years ago, or maybe even a year ago, because each time, the improvements are likely to reveal more new things we missed before.

 

If I may I would like to show just one small example of what this combination of technology and software developments can now do.

 

This is a very high speed video of detonation cord exploding - to give you some idea the detonation is moving at nearly 8 km/s. It is only a frame rate which might have been unbelievable a few years ago, which makes this possible at all. The link and reference to the original video can be found in the you tube description for this.

 

What we did was to transform it by running it through very many "change detection" algorithms to try and highlight something we saw very faintly in the original - the crossing of the air shocks driven out after the cord has detonated [because this has important learning outcomes for blasting engineers in terms of visualizing the shock wave processes]. This video shows the results for three particular attempts, the third one shows up what the eye cannot see in the video very clearly, note the air shocks propagating out left and right as white lines.