Military vehicles and soldiers are being equipped with a greater number of sensors for threat detection in an effort to keep soldiers safe – the SPIE Defence, Security and Sensing show in Baltimore, USA, which ran from 5 to 9 May, showcased some of this sensing technology. But with several countries, including the UK and USA, recently announcing budget cuts to the armed forces, cost is becoming much more of a concern – adding an extra factor to existing requirements of reduced size, weight and power consumption for imaging systems used by the military. Furthermore, with the move to increase sensing capabilities onboard military vehicles, there is the additional challenge of how to make sense of all the data.
Several industries often require fast detection – but, for the military, the speed of detecting threats such as explosives or sniper attacks plays a fundamental role in protecting the lives of soldiers.
A muzzle flash – the burst of light from the muzzle of a firearm when it is fired − can be used to identify the presence and the direction of sniper fire. High-temperature, high-pressure gases from the muzzle emit an infrared signature, but the flashes are over so quickly that they are often difficult to detect. ‘Normally, muzzle flashes would be so small and so fast that a normal infrared detector wouldn’t see it − it would be gone before it [the detector] had a chance to detect it,’ said Piers Rivington, business development manager of the lasers and detectors division at Pacer International.
The Tachyon series of high-speed mid-wave infrared (MWIR) lead selenide detectors supplied by Pacer are capable of responding very quickly to infrared markers such as muzzle flashes. ‘It sees the IR signature emitted from the very hot gasses coming out of the muzzle,’ Rivington explained. ‘Because these detectors have such a fast response time, [the detector] is able to see that as a transient flash or a very fast IR signal.’ The speed of the detector is due to the way in which the lead selenide is deposited, according to Rivington: ‘The lead selenide is deposited by vapour phase deposition, which allows a very thin layer of material to be deposited and formed onto the detector.’
Infrared sensors have been designed to detect sniper fire, but also other threats such as explosions. ‘One of the potential applications has been defence against improvised explosive devices (IEDs), because you need detectors that are fast enough to detect an explosion very close by, and react fast enough to deploy some sort of countermeasure,’ commented Rivington. In addition, this type of sensing technology can be used to characterise the explosion or detonation and find out details of how certain IEDs work in order to protect against them in the future.
Another development taking place within military and defence that Rivington noted is the use of IR signatures to help identify soldiers in the field. The soldier would wear an emitter that would identify them as a friend, as opposed to an enemy. ‘The sophistication of the technology that is attached to individual soldiers is certainly going up, and using IR emission to identify individual people − in friend-or-foe situations − is a trend,’ he said.
Depending on the threat will determine how many and where the infrared sensors are positioned on a military vehicle, for instance. ‘For muzzle flash detection you would need several detectors spread over a military vehicle,’ Rivington remarked, to cover the widest field of view possible. ‘But with an IED detector, you would need far fewer of them because the explosive would be much closer to the vehicle. And it would be a completely different signature altogether.’
Dealing with data
With multiple imaging sources being deployed in the same systems onboard military vehicles, there is a growing need to advance video distribution capability through Ethernet, so that all personnel, not just the driver, can view and share images and video on any display. But with cost being an issue, there is hesitation to replace existing systems to achieve this. ‘Most of the cameras tend to be very expensive and specialised, and in many cases the lenses and the optics are often more expensive than the cameras and video equipment themselves,’ explained John Butler sales manager at Pleora Technologies. Instead, external frame grabbers can be used to add analogue and Camera Link cameras to a network without having to replace cameras or design adapters from scratch.
‘With external frame grabbers we provide defence integrators and system manufacturers with the ability to convert the image feed from existing cameras into a GigE Vision stream that is transmitted alongside video from newly installed, high-resolution digital cameras on the same Ethernet network,’ said Butler. ‘With all the video in a common format and on the same network, it’s easier to broadcast that video to various display screens and processing computers throughout a vehicle.’
A potential problem, however, with having more people viewing information on displays is that the images may not be easy to interpret for military personnel who are not trained to read them. ‘I think for soldiers, the near infrared is easier to interpret compared with thermal images that are more blurred and tricky to interpret. For example, if you image a person with thermal imaging it may be quite difficult to recognise this person,’ commented Dr Svensson of the Swedish Defence Research Agency. ‘We [Swedish Defence Research Agency] are developing methods to make images in the thermal spectral range easier to interpret for a military operator.’
As witnessed by Svensson at SPIE DSS, in the defence space there has been an upwards trend to combine different imaging technologies to increase capability: ‘If we look at the event in Baltimore, we see a big trend to fuse different technologies − for example image intensifiers with thermal infrared,’ he noted. Pleora’s Butler added that this means that more sensors are being installed into military vehicles: ‘In some cases, where you would have had four sensors, you may have eight today because you might be putting visible and thermal cameras together, where previously you only had visible cameras.’
The increase in sensors means that onboard computers have to be able to process much more data. ‘The amount of data that has to be distributed in the system is going up, and it’s increasing aggressively,’ commented Butler. And, the desire for higher resolutions and speeds raises this further. ‘If we step back a few years, a lot of the older legacy interface technologies were adequate for transmitting the required video resolution, the frame rate, and so on. We come up to the present day, and there is a growing drive to be able to use higher resolution sensors, faster sensors − so your frame rates are higher,’ Butler continued. ‘Keeping pace with data rate requirements is going to continue to be a growing challenge as imaging applications become more complex.’
The real-time capability and fast response times needed for the detection of muzzle flashes and explosives, for example, also drives up the data rate. Using the example of the Tachyon MWIR detectors, Pacer’s Rivington outlined the sheer volume of data that such military imaging systems have to deal with: ‘It has a detector array with 80 x 80 pixels in it, which is fast enough to scan at 1,000 frames per second,’ he said. ‘That is an enormous amount of data.’
One way to manage the increasing quantities of data is to use faster connectivity solutions. Pleora’s Butler commented: ‘In the defence space we are starting to see more interest in using 10 GigE, which allows the movement of a larger aggregate data rate. So, if you’ve got a number of different cameras you can consolidate multiple GigE links into a 10 GigE stream.’ This would also accommodate the faster and higher resolution sensors being deployed, such as those described to detect muzzle flashes. ‘As the military uses these higher data rate sensors, 10 GigE gives them the bandwidth to distribute that sensor data without sacrificing frame rate or resolution,’ Butler said. ‘Data rates are starting to exceed 1 GigE on a frequent basis for many military imaging applications.’
Another way in which high data rates are being processed more efficiently is through the use of embedded processing. ‘The Tachyon detectors have the read-out electronics embedded into the same chip as the detector,’ Rivington explained. ‘That is one way of dealing with the enormous amount of data generated.’ Pleora’s Butler agreed that embedded processing is used, and will be used much more, by the military. ‘I think you’re going to see a lot more activity towards embedded processing capability,’ he noted. ‘The embedded processing world is on an upward trend in terms of the amount of performance you can get out of those systems. It is really becoming attractive for these kinds of applications now, because they are smaller, lightweight and power efficient.’
Cuts in defence budgets mean that cost is more of an important factor than before, and so the military is using systems they can purchase off-the-shelf to speed up the integration of new technology. The wider adoption of machine vision standards and products across other markets could also translate into cost-savings for military imaging applications. ‘I think the defence industry has been doing the right thing in trying to leverage standards from the industrial machine vision market – that has significantly opened up the range of products they can buy off the shelf,’ said Butler. ‘There is a lot of high-quality standards-based imaging equipment that is deployed or being developed for industrial, transportation, and medical purposes that can be repurposed for defence applications.’
In terms of the future for imaging technology used by the military, both Butler and Rivington suggested that soldiers will start to carry more imaging equipment. ‘There will be a continued direction towards unmanned or dismounted systems, and that will lead you into things such as mobile technology, cellular capability, where you are going to want to put all of that processing and imaging equipment on the soldier,’ Butler said. But system size, weight, power efficiency and processing capability will have to be improved further to allow this to happen, he added: ‘You can’t do it with the bulk and the weight of the equipment today − it’s going to have to get smaller and lighter and they [imaging equipment] are going to have to be able to operate for extended periods of time on batteries than can be man-carried.'
