Sea view

If you buy a camera today, the chances are it will be digital. Film is almost a thing of the past, with sensors using charge-coupled devices providing high resolution and compatibility with computers. Now, digital photography has found a niche in a realm that would challenge even the most daredevil snapper: underwater holography.

A team led by Professor John Watson of Aberdeen University’s department of engineering has developed the world’s smallest underwater holography camera, the eHoloCam. The camera has just been tested on a series of research cruises and is likely to be used to provide valuable information on the behaviour and environment of plankton and other marine species. It could also be used to study offshore engineering structures, giving a more complete picture of how they affect the seabed sediments.

‘Looking at undersea life wasn’t initially the driving force for the project,’ Watson explained. ‘I’ve been working on this sort of stuff for 30 years, and I originally started thinking about it for inspection of oil pipelines. But that got superseded about 15 years ago, when we realised there were real subsea biological problems that holography could solve. Since then, that’s where we’ve got most of our funding from.’

Understanding the location, distribution and behaviour of plankton species increases understanding of the ecology of the oceans and how plankton influence the health of fish stocks, Watson said. Holography is important to marine biology because it allows researchers to see plankton or other organisms in relation to each other and the particles of silt and organic matter surrounding them.

‘A straight photograph, even a stereo photograph, loses the parallax information, so you can only know where an object is; it can’t tell you anything about its relationship to anything else. It’s that spatial information which is very important to marine biologists and workers in other fields as well,’ Watson said.

‘The other reason is that because it depends on optics, it’s a non-disruptive, non-invasive technique.’

Watson’s first underwater holography equipment was called the HoloMar. Developed in the 1990s and deployed in 2001, it was a monstrous piece of equipment, measuring 2m by 1m and weighing 2 tonnes.

‘It was a bit unwieldy and not exactly user-friendly,’ said Watson. ‘But it had to be that size, because it needed to incorporate a large laser, physically large and very high-energy, and it recorded on holographic plates.’ Analogous to the light-sensitive plates in an extremely old-fashioned camera, these were sheets of glass treated with light-sensitive chemicals and were heavy and awkward to handle.

The eHoloCam, by contrast, is a cylinder 90cm long and about 30cm in diameter, and weighs 75kg complete with mounting frame. It consists of a housing, containing the laser, power supply and associated electronics. The laser emerges from a lens on the end and shines towards a 1cm diameter sensor, which is mounted on an end-plate 50cm away on three rods.

‘The concept of digital holography itself — using an electronic sensor to record the image and processing the data electronically — was what helped us to make the camera so much smaller,’ Watson said. ‘And what undoubtedly helped was the fact that you can get much higher power in more compact lasers than when we developed the HoloMar.’

The job of designing the eHoloCam laser went to Elforlight, a specialist firm in Daventry, Northamptonshire. ‘If you want to make a hologram, you need a coherent light source,’ said Keith Oakes, who designed the laser system. The electronic image recording process requires less intense light than the old holographic plate system, so the laser can be of a lower power, and therefore smaller, he said.

The eHoloCam passes light through the semi-transparent plankton. When it emerges on the other side, its wavelength is shifted slightly. This combines with the light passing either side of the plankton and forms an interference pattern, which is recorded by the sensor. The wavelength of the light from the source needs to be constant for the entire recording length. This is known as coherence, and ensuring the laser produced 50cm of coherent light was one of the trickiest problems for Oakes and his team.

Oakes solved this by careful design of the laser cavity, the component whose optical characteristics generate the laser light. In this case, this was a solid crystal of a compound called yttrium aluminium garnet, doped with neodynium (known as Nd-YAG, a common material for solid-state lasers). ‘In the end, we managed a coherence length of 80cm,’ he said.

The laser also had to produce a light wavelength capable of penetrating the full length of the 50cm column of water between source and sensor. ‘Blue-green wavelengths are best for penetration of the water,’ Oakes said. ‘The Nd-YAG laser produces a wavelength of 1064nm, which is in the infrared, so we used a non-linear crystal to double the wavelength into the green range.’

Neither the laser nor the eHoloCam are the finished article. ‘We started off developing a version which would include both in-line holography and another system, off-axis holography,’ said Watson. Off-axis, which was included in the HoloMar system, bounces laser light off the object rather than passing it through. This allows the system to take pictures of larger, opaque objects, such as agglomerations of sediment particles or bigger organisms, including fish.

‘Then we were told that if we were quick, we could take the instrument on some trial cruises on the RV Scotia, a research vessel operated by the Fisheries Research Services’ Marine Lab here in Aberdeen.’

This meant Elforlight also had to curtail its development time. ‘Originally, this was being developed as a benchtop prototype and another laser, a further development, would be deployed underwater,’ said Oakes. ‘But the opportunity of the research vessel, with all the crew and equipment and so on, was too good to miss, so we did a few tweaks to the engineering to ruggedise it.’ The components were then sent to an offshore engineering specialist, CDL in Aberdeen, which designed and built the watertight housing to package the camera.

The rapid development and use of a prototype means the laser is oversized and overpowered for the job it does in the eHoloCam. Watson’s team is continuing the development using the same laser to provide off-axis holography within the same housing as the eHoloCam. ‘But if it was a wholly in-line system without the provision for off-axis, it could be smaller yet,’ Watson said. A version using the lower-powered laser needed for in-line only would be about half the size, he estimated.

The team is still grappling with the data processing. ‘The hologram is a diffraction pattern, which contains an enormous amount of information,’ Watson said. ‘We then take that information and by various software algorithms, reconstruct the holographic image in planes 1mm thick inside the computer. You then tell the computer that you want to see the image at a particular distance from the sensor, and it reconstructs the image in that plane. You don’t see the whole lot as you would if you were looking at a physical, photographic hologram. You navigate through the image. We’re still working on the problem of being able to extract and analyse all the data and display it efficiently and quickly.’

The trial cruises, where the camera was trailed behind the RV Scotia on a submersible frame, were hampered by the weather. The plan was to submerge the camera to a depth of 1.5km, but it was only possible to go 500m down. However, the results were extremely encouraging.

Oakes hopes they have attracted enough interest from the biological and offshore research communities to attract further funding. ‘The group at Aberdeen are excited about the biology that’s been revealed, and it can also gather information about deposition of silts and movement of material underwater: that will help in real-world engineering,’ he said.

Oakes hopes to continue development and Watson is continuing to work on the full-scale, twin-axis original concept and a smaller, inline-only version. Both would include extra control systems so the cameras could be mounted on remotely operated submersibles. The smaller systems would be cheaper and might be sold to research organisations in the biological and offshore engineering sectors. The larger system, incorporating off-axis, would be more expensive and might be hired out.