Q & A with a Library photographer
A look at using Reflectance Transformation Imaging (RTI) to uncover our collection
19 August 2016
With all the projects and busy day to day work of exhibitions, sometimes its nice to step back and chat to someone about what they do. When you work in an organisation as broad and varied as the Library, there is always so much going on you don't know about. We all have pet interests and new ideas about our work but don't get alot of time to talk about them, so let's hear from photographer Craig Mackenzie about what he's up to.
What do you do at the Library?
I am a photographer. It's my job to digitise the Library's collections—make sure they are accurate digital representations of the original painting, document, object or photograph.
How long have you been working at the Library?
I came to the National Library of Australia almost 10 years ago after studying photography at CIT for 3 years. Prior to this I had a great variety of jobs including furniture removalist, factory worker producing industrial pumps, fruit picking, cleaning, ground staff at Royal Canberra golf course, abattoir slaughterman, specialist timber salesman, picture framer, furniture making and many other bits and pieces.
What was the best shoot you’ve done?
I don’t think there has been one best shoot as such. However, projects commissioned by the Pictures Collection are memorable, and although they are often quite demanding in a logistical and technical sense, there is a sense of achievement and satisfaction when the portfolio of images has been selected.
What are you working on now?
I am currently shooting a selection of Japanese Woodblock prints for an upcoming exhibition. These prints have already been digitised in a traditional manner, but I am shooting them using a new technique called Reflectance Transformation Imaging (RTI). This technique provides curators, conservationists, collection managers and scholars a new way of viewing and understanding the works they are researching.
Explain to us how RTI works
Reflectance Transformation Imaging involves taking a series of photographs where the object and camera do not move in relationship with each other, but the direction of the light source moves for each shot. For these prints I am taking 48 images, moving the flash in each image. It is probably easiest to imagine a dome of lights above the object and each light position being placed on one of 12 points of a clock face, at an angle of 15, 30, 45 and 60 from the horizontal. There are special spherical targets within the frame of each shot which reflect the specular highlight of each flash allowing the software to calculate the direction the light came from.
When these 48 images are processed, the RTI software is able to create a 3 dimensional model of the surface of the object. Works like prints that we consider to be 2D, are transformed into 3D surfaces with hills and valleys where the woodblock made its impression on the paper, or where the artists etched a pattern into the ink. This can then be manipulated to extract data about surface texture and structure not immediately obvious or even hidden.
How did you learn about this technique?
I attended a workshop at the Grimwade Centre at the University of Melbourne. It was run by an organisation called Cultural Heritage Imaging. They have been at the forefront of this developing field for several years now. It was an intensive four days of training involving other photographers, conservationists and even an archaeologist.
When you first heard about RTI, where did your brain take you in terms of the possibilities?
When I first heard about this technique I was curious. I could see that it was another way of documenting collection items but it wasn’t until I had completed the course and seen the many examples that its potential was really understood. The number and type of collection materials the Library holds which might benefit from this technique seems to grow every time I think about it.
Has it changed the way you think about the role of photography in cultural heritage conservation?
Photography has always been important for documenting collection items for a variety of reasons. RTI and other computational methods are exciting in that they not only document an item but can be used to reveal otherwise hidden details and extract data that has not been previously available.
Is there a whole lot of really complicated mathematics involved?
The theoretical basis of RTI is quite complex but from a practical point of view it has been designed to be very user friendly. It is possible for anyone with sound photography skills and a little training, to set up, shoot an object and get usable data.
What do you think the biggest benefits of this kind of documentation are?
This is a powerful tool which adds to the techniques already available to study and monitor objects and artworks. As such it will add to our understanding of these works, how they were produced, reveal hidden detail, record physical condition and track changes over time. For example, a researcher 10 years from now will be able to pick up these files and map them to their own images to see if the varnish on a painting is deteriorating or if a tear has gotten worse. It is also very easy to share files so others can collaborate and consult. The ability to see and manipulate the files for themselves allows others to explore the one item in different ways and bring their own knowledge and experience to the object being looked at.
Why is it important to capture this kind of information? What will it mean for future researchers?
This information will benefit conservationists, curators, collection managers and scholars in their research and study. The field of computational photography is developing quickly and in the future it will be possible to reprocess the image files shot now to produce new information and insights. As long as the initial series of photographs is captured to a high technical standard, the requirement of reshooting and all the handling which goes with that will be eliminated.
If you could pick one object/site/space to use RTI on, what would it be? And why?
This is a tough one. Off the top of my head and having done no research to explore the idea yet, I think it would be pretty cool to document Aboriginal rock paintings. Technically it would be challenging, but I don’t think it has been done before and it would be interesting to see what would be revealed.
Thanks Craig, can we check in with you in a few months and see how it goes?
Sure, I look forward to it.
And finally – just because we are a library – what are you reading right now?
‘The Great Divide’, by Joseph Stiglitz and ‘Celestial Navigation is Easy’, by Bryan N. Cox