Episode 11: Multispectral Imaging

“I did not expect to see a King riding an ostrich on a mountain in North Africa.”

Available on Anchor, Apple Podcasts, Breaker, Castbox, Google Podcasts, Overcast, Pocket Casts, RadioPublic, Spotify, Stitcher, and TuneIn.

About this Episode

Recorded August 2, 2021. Produced and edited by Hannah Busch.

Guests: Katie Albers-Morris, Dr. Helen Davies, Alex Zawacki

Content: In this episode Katie Albers-Morris, Helen Davies, and Alex Zawacki talk about recovering palimpsests and erased texts with multispectral imaging. All three are, or have been, PhD candidates at the Lazarus project at the University of Rochester, an initiative that was designed with the educational purpose of training students in the field of multispectral imaging and image processing techniques for cultural heritage objects. During the episode we discuss MSI in general, their experiences as (grad) students and program coordinators at the Lazarus project, MSI in the classroom, and the challenges of dissertation projects in the digital humanities.

Further Reading/Resources

Episode Transcript

Automated transcription by <https://otter.ai>, manually corrected by Hannah Busch.

[Music Fade In]

Hannah Busch 0:11
Hello and welcome to Coding Codices a podcast from the Digital Medievalist Postgraduate committee. I’m Hannah Busch, your host for this episode, and I’m joined today by Dr. Helen Davies, Katie Albers-Morris and Alex Zawacki. In today’s episode, we are going to talk about multispectral imaging, and imaging and image processing technique that allows to recover contents from damaged materials.

[Music interlude]

So to get started, I would like to ask you to introduce yourself quickly to the audience.

Helen Davies 0:44
I’m Helen Davies. I’m an assistant professor of English at the University of Colorado, Colorado Springs, and I was hired to teach a combination of the Digital Humanities and medieval literature. So I get to combine both of those things in my classes and my work. And I am working on introducing multispectral imaging to the University of Colorado, Colorado Springs. I have previously worked for six years for the Lazarus project as a project coordinator.

Katie Albers-Morris 1:17
And I’m Katie Albers-Morris. I am a PhD student at the University of Rochester, and Project Coordinator for the Lazarus project. I am working with the Lazarus project on cultural heritage imaging as we do, but my specific interests are points of cultural intersection in the Mediterranean. So where the Islamic world and the Christian Latin West kind of meet, etc, and the role of women in those spaces as well.

Alex Zawacki 1:46
Hi, I’m Alex Zawacki. I’m a PhD candidate at the University of Rochester where I work with Katie at the Lazarus project. So my research also focuses on the Middle Ages, specifically the supernatural, ghosts or that sort of thing. And I’m an Operations Coordinator with Lazarus where I’ve helped run a couple of projects in the past, including the Museum of Holocaust in DC, in Dresden, and a couple of other places trying to recover damaged palimpsests and medieval manuscripts.

[Music interlude]

Hannah Busch 2:21
Yeah, I’m familiar with MSI. And I think I got introduced to it the first time via the Archimedes Palimpsest. But I’m not sure if all the listeners of the podcast know what MSI is. So would you would you explain it? Short and easy and understandable?

Helen Davies 2:40
Yeah, so the short and simple version is, if a document is illegible because it’s faded, because it’s damaged, because it’s been through fire, or in dramatic instances, bomb or flooding damage, then we can use multispectral imaging to recover the text. Additionally, like as you mentioned, that Archimedes Palimpsest project, it’s frequently used to recover the under-text in a palimpsest, so if a text has been erased, scraped off, etc, we can use multispectral imaging to try to recover that text. So the technology works by taking a series of images under controlled bands of light from the UV through the visible spectrum and into the infrared. And then we can combine all of these images together to create the data cube that we extract the information from. So we use various imaging software that can then recombine these images, and provide more information than any single image would be able to on its own. And it can let us read these texts that are otherwise lost to us.

Hannah Busch 3:47
And what specific tools do you need for this? You mentioned software and also, I guess the hardware?

Helen Davies 3:55
Yes. So there’s a couple of different companies that make the multispectral imaging camera systems and then there are a few, well, there’s one a recent NEH grant to create a simple and portable and inexpensive multi spectral imaging system. So a multispectral imaging system will consists of a camera, the lights, and a special type of lens. And so the lights are narrow-band LEDs that we can control, specifically what wavelengths are being emitted. Then there is a high resolution camera and a lens that doesn’t distort the light when it’s in the uv, the visible spectrum, or into the infrared. So most camera lenses, for instance, the one on your iPhone, filters out for certain bands of lights and also will distort outside of the visible spectrum. These specific camera lenses don’t do that. They are, they are designed to be sharp outside of the traditional range of light that can be captured by a commercial camera. And then as you mentioned earlier, you need special software. So there are a number of different software packages that can work with multispectral imaging, including ENVI is the one that was primarily used by the Archimedes Palimpsest guys, originally, I think, and then a lot of them also wrote their own software. So Keith Knox, who wrote, who worked on the Archimedes Palimpsest project, has recently written a new software called Hoku, which is open access, so you can use that. And then Dr. Tania Kleynhans at the Rochester Institute of Technology has also recently written an open access multispectral imaging, image processing software, as well.

[Music interlude]

Hannah Busch 6:00
You all mentioned the Lazarus project. And I looked it, like when I looked it up, I was super hooked that you actually include students and that you teach MSI, and that students are able to work with it. And I think that’s a pretty unique initiative. Can you tell me something more about the project, or?

Helen Davies 6:19
Yeah, so Dr. Gregory Heyworth started the project when we were at the University of Mississippi. And when he transferred up to the University of Rochester, he brought the project with him. I will let the two of them speak to more recent projects as they still work for them. But Greg designed the Lazarus project to be a training project. So it was designed with an educational purpose in mind. And so the Lazarus project has traditionally worked with a number of undergraduates as well as graduate students training them in different aspects of multispectral imaging. And I will say, having worked with a number of undergrads, as well, as you know, getting to work with Alex and Katie on the Lazarus project, that one of the cool things is that everybody, by bringing a bunch of different students onto the project, we get to work with a variety of people from different backgrounds that all bring some kind of new perspective and new skills to the project in a way that’s really fun and really interesting. You want to add to that, Alex?

Alex Zawacki 7:21
Yeah, just from the beginning, Greg’s goal for the projects for the Lazarus project have been that it be pedagogical, as well as focused on recovery. So it’s not strictly focused on producing product and publishing results and the like. It’s also focused on training undergraduates and graduate students, and giving people the experience to learn to work with these materials. And towards that end, unfortunately, we’re limited by the number of people who can work with the study one time because our lab can only fit so many human beings in it at once. But towards that end, we’ve had we’ve brought on a lot of undergraduates who’ve done with everything who’ve helped us with everything from writing apps and programs to doing image processing to working with manuscripts themselves, which I think has been a really good experience for us because it is constantly broadening our horizons by bringing in students from different backgrounds, and for the students by giving them the opportunity to play with these things.

Katie Albers-Morris 8:16
I would also add that I was an undergraduate who worked on the project back in Mississippi, and I have since obviously progressed to continue working with Lazarus project still learning all the time. And I’ve seen things from, you know, a student building a neural net to read medieval hands, right? All the way up to 3D Image Capture we did last semester in a class we were teaching. So it’s a really great kind of moving forward with technology kind of initiative that I think really helps students who are interested in the tech but maybe are more conventionally trained like myself, or vice versa. In the case of Helen, it really helps to bridge that gap between the digital and the humanities,I think.

Hannah Busch 9:06
That’s so great. But yeah, what is about, I was wondering about the general background of the students from which departments do they come? How did they get introduced into the project?

Alex Zawacki 9:17
Well, that’s forgetting introduced to it. I think any one of us who has taught a class or TA to class has given some kind of spiel on ourselves, which has involved talking about the Lazarus project. I know I’ve done that for the classes I’ve taught just as in the context of introducing myself, but there’s also the classes that Greg teaches, including there’s a DMS course, that focuses on the interaction between digital and physical, the ideas in of mediating the physical through the digital, which has a really broad background, which brings in students of really broad background from everyone from film students at the graduate level, English students, various students from the sciences, and that class has introduced a lot of students to the Lazarus project because a big component of the class is working with the Lazarus project. So students there will work on multispectral imaging and object. We’ll talk about other forms of imaging. They’ve done things like structured light scanning and photogrammetry. Really anything that we have the technology and the capability for them to get hands on experience with, they will do it in the course of that class. And Helen and I have both at various times TA that class and Katie, did you ta at this past year? Yeah.

Katie Albers-Morris 10:29
Yeah, I just did it this past semester in the spring.

Helen Davies 10:33
But yeah, to build on what Alex was saying. I mean, we’ve had engineering students who come and they want to design a new physical system for it. We’ve had comp science students that bring a new understanding of a new processing algorithm, as well as English and History students that were like, Yo, but like, check out how cool that manuscript is is. And this summer, I’m teaching at UCCS. I’m teaching a similar course as the DMS course that Katie TA for last semester, and that Alex was talking about, and like, I have a bunch of pre-med students, because it’s like, they’re trying to make connections with it with medical imaging, like and their understanding of it. So yeah, so you get students from all different walks of life.

[Music interlude]

Hannah Busch 11:18
The hardware you need for multispectral imaging is pretty expensive. But I remember your talk, Helen at Leeds, that you talked about this new system that is more affordable and also easier to transport. And would you would you mind to speak a bit about that?

Helen Davies 11:36
Yeah, so Dr. Tania Kleynhans and Dave Messinger at the Rochester Institute of Technology recently wrote an NEH grant to fund the development of a new super portable and affordable multispectral imaging system. And so it will be about $5,000, which is significantly more affordable for people like me, who teach at a state institution, or perhaps smaller libraries and archives. And so it also opens up and expands the types of documents that we can work with, where we’re no longer focusing on boutique imaging projects, and instead can work with things like an Ellis Island diary, or you know, letters is that may or may not have interesting information. The thing about the hardware, though, is that this new portable system has a lot of potential promise. And there’s still basically going to be a niche for both styles of systems. So the new more affordable portable system will be able to work with smaller institutions that can’t afford the larger system. But then the larger system, the more expensive system will be able to do that higher end projects for the more intensely damaged documents. Or, if there is information that the newer system might not be able to recover, then the more expensive system, we’ll be able to do that. So they kind of can fit together in really interesting ways to cover a wide range of different types of damage, different types of information that can be recovered.

[Music Interlude]

Hannah Busch 13:15
I was wondering, like Rochester, I think it’s not very famous for its medieval manuscripts. So I was just wondering, what kind of documents are you working with? I know that you also have traveled a lot, but I would be curious to hear about the kinds of objects you’re working with, on location. And in Rochester.

Alex Zawacki 13:37
So actually, one of the first things that the Lazarus project imaged at Rochester, it was actually sort of the the demo, the objects that Gregory gave when he first came to the university was a damaged manuscript fragment that’s been in the university’s cares since the 60s, I believe. And it’s in really poor shape. It was used as a, it’s a, it’s a binding fragment. So it’s got sewing holes down the middle, it’s almost all the text is worn away. It’s very badly stained mess. And that was one of the first things that Lazarus worked on recovering at the university. We ended up finding out that it’s the text of a 14th century copy of Richard FitzRalph’s Summa de Questionibus Armenorum, which is the theological text essentially about the difference between the Armenian Church and the Catholic Church, and it’s the only copy of it that exists in a non European library. And it’s also, so argues our paleographer, who’s much more knowledgeable about these things than I am probably the oldest copy of it that exists. So that was quite a fun find. So the university does have the Robbins Library, which is a medieval focused non lending library which is one of University of Rochester libraries. They do have some medieval manuscripts with the in the last couple of years there’s been more of an emphasis on acquiring new One, so they are slowly building a larger collection. But most of the work that we’ve done has been traveling for those purposes because we as you said, we don’t have a huge collection. Although we have some interesting things that we have to image in the coming days, like there’s a medieval library scroll that needs to be imaged, there seems to be some erasures on that. We’ve also done some imaging for the community. So we had a very sad, it’s somebody whose parents when her father had written a Valentine to her mother, both of whom had passed away on a chunk of linoleum from the home that they had owned together, which then burn down and most of the texts on it had been rendered illegible by the heat, it hadn’t burned directly, but it had been damaged. And we managed to recover that for them. And we’ve had some other community members bring in some more things like that. So there are some things in the university itself and then there’s always kind of an ongoing dialogue with the community to bring us things that we can help out.

Helen Davies 15:57
Alex, don’t forget the one manuscript that the Smithsonian popped in the mail to us.

Alex Zawacki 16:02
That’s right, yes, every now and then someone will FedEx us a manuscript. So we imaged a manuscript for the Smithsonian. It was, and actually, that’s an ongoing project, right now we’re working on deciphering it, they have a really interesting, very small, extremely fragile, 14th century Armenian prayer book with the original bindings. It’s very, very delicate, it was very nerve wracking to image it. And so they they sent it to us to image at the university rather than our going down to DC to image it there. And that’s, that’s we’re working on processing and deciphering that at the moment. But so every now and then someone will just ups, a priceless document to us.

Helen Davies 16:44
And that one was particularly interesting, because it did allow us to work on it in the University of Rochester in the library or in our lab, and it took several weeks. So it was quite good that it had been sent to us because that would have been quite the commitment to go down to the Smithsonian to image that consistently. The other things that I just want to mention really briefly, is that Dr. Anna Siebach-Larsen has been working on acquiring new medieval manuscripts for the Robbins Library that Alex mentioned. But she’s also focused on acquiring damaged manuscripts. Because the Lazarus project has the ability to recover them. So then it can work hand in hand with the university and the library.

[Music interlude]

Hannah Busch 17:35
Yeah, if you think of MSI, I think most people really think about palimpsests and damaged texts. But are there other options, what you can use MSI for. So I think if someone gets the option to do MSI on their collection, but maybe they don’t know if they even have something interesting. Can you give some tips or what material can you work on? And what is promising? Just to give any direction to ,yeah, to people who might be interested in doing MSI?

Helen Davies 18:05
So I’ll let Alex speak about, or Alex and Katie speak about, the medieval manuscripts that they are working on recently, the different types that allows our strategic is working on. Hannah, as you mentioned, like we were talking at Leeds about some of the weirder documents that I’ve been working on this summer. And so I have, you know, the technology was developed for the Archimedes palimpsest and is primarily been used up until this point on a lot of medieval manuscripts. Because I’ve been working with a portable system, because I’m working in a location that doesn’t have a whole ton of medieval manuscripts. I’ve been working a lot on documents that I previously had no experience working with multispectral imaging on. So I’ve been working on an 1800s Ellis Island diary, some letters from the 1800s. There was, one of the weirder things is, there’s a old mining camp in Colorado that I swear to god is called Buckskin Joe, and it’s one of the first mining camps in Colorado. And we the law code survives. And the law code is currently at the Colorado Supreme Court Law Library. However, you can’t read it anymore, because it’s pretty faded. And so one of the documents that we were imaging this summer was this Buckskin Joe Legal Code. And we’ve been able to recover that. So it doesn’t have to be just medieval manuscripts. It can be a wide range of things, including kind of more common documents that one might find in America.

Katie Albers-Morris 19:37
I was also thinking about when we were at the University of Mississippi, they found a fingerprint on one of Faulkner‘s like kind of personal copies of something. And so it became a whole long is this actually Faulkner’s fingerprint like have we found this, you know, very tangible piece of him and of course, Faulkner in Oxford, Mississippi is a big deal, his house is there, and you know, every grad student has to sit through at least one class about Faulkner, it’s just kind of the way it is. So, definitely things like that. And also, just last semester, we were working on a manuscript, a bit of sheet music that Fanny Mendelssohn wrote, and a student was able to see kind of through the paste down, where she had, you know, glued something down and changed what she wanted to write, but also where she had crossed things out, we were able to see a little more clearly the original direction and music before she changed it.

Alex Zawacki 20:39
It’s also worth mentioning real quick and in terms of addition, in addition to recovering things like erased text, or damaged images, or kind of our standard stock and trade. There’s also a lot of potential for applications and conservation of any kind of document with a medieval or modern or classical, pick an era. Because degradation tends to show up in the infrared, invisible portions of the spectrum, especially the infrared before it becomes visible to the naked eye. And so one of the things we’re exploring right now is imaging separate documents, separate paintings over long periods of time, in a full spectrum. And trying to track when that kind of degradation appears and how you can predict it, and how you can track it using MSI, prior to it showing up to the naked eye. So that you can get kind of a jumpstart on on being aware of problems with your documents. So we think there are going to be some pretty significant avenues for exploration in conservation, in addition to strictly recovery.

Hannah Busch 21:39
So I was working a lot on, or I’m still also kind of working on manuscript digitization. And I was wondering, like, it would be amazing to of course, have multispectral images for every manuscript that you digitize. But of course, that’s probably not affordable time and money wise. Do you know of any projects, except for the more famous ones like the Sinai manuscripts, where this has been discussed to include multispectral images to the digitization process, or have you thought about it to offer images to the collections.

Alex Zawacki 22:21
So I know the British Library, for example, has a multispectral imaging system run by Christina Duffy, who is wonderful. They, as far as I know, aren’t digitizing things with multispectral imaging as a matter of course, unless there’s a particular reason to do so. So, in the way that you can request images of manuscripts in their collection, I believe you can request multispectral images as well. Although they haven’t, they haven’t applied to kind of standard digitization scheme yet simply because they have so many manuscripts and one system and one person running it. So I think that might be more of a thing in the future is institutions offering kind of bespoke multispectral imaging where you can request it. I’m not aware of anybody, besides the ones you mentioned, like Sinai, who are digitizing and applying MSI as a matter of course. Helen or Katie might be able to jump in on that.

Katie Albers-Morris 23:15
I know that relatively recently someone has reached out to us about this, I don’t necessarily want to say too much, because I don’t know how far forward it’s going to go. But there definitely have been relatively recent inquiries into having a partnership with digitization and also multispectral teams. So if that does move forward, we would probably be working alongside the early manuscripts electronic library team to do the digitization and the multispectral imaging of a collection.

Helen Davies 23:49
Similarly, I know that one of the hopes with the more affordable, the RIT system, is that it can be used, especially because it won’t be for the highest end or the most challenging documents that it can be used as kind of not quite mass digitization, but in conjunction with more traditional digitization or get a wider scale. So Gregory Heyworth at the University of Rochester and I had previously discussed that one potential avenue is that if the Lazarus project is imaging a number of really damaged documents, then perhaps we could use one of the RIT systems to kind of get a broader spectrum of the rest of the medieval library that may exist in a certain location, so that you can kind of get the boutique imaging of the high end, the superstar documents, but then you can also get this wider range. And then one other thing I want to add is tha,t so, I’m super grateful for two for example, Colorado College for letting me borrow their medieval manuscripts for the summer to work on multispectral imaging when my own university doesn’t have any special collections. And so in return, then I’m going to give them all of the multispectral imaging files, as well as, you know, the pictures that we took of them so that if they want to host them, in addition to the new digitized images, they can put that all up on their website. I will say that multispectral images, the data files can be robust. And so they can take up a lot of space if you’re thinking about hosting them and alongside traditional digitized snapshots of a manuscript. So, it might be the kind of thing that as the digitization process evolves, and as our work with multispectral imaging continues to change and shift, you may be able to produce more mass kind of digitization efforts, including multispectral imaging, but hosting them might still be a bit of a bit of an issue, a bit of an effort. And so that’s just another aspect of it to consider going forward.

Hannah Busch 25:58
I know that multispectral imaging is a non-invasive technique, but, and I also have a feeling from what you tell that in your surroundings, everyone is super open and really interested in collaborating with you. But I think that that might be also collection holders that are afraid of this imaging technique. So can you tell us a bit about the invasiveness or non-invasiveness about multispectral imaging?

Helen Davies 26:25
Yeah, so I mean, again, I’m sure somebody else can jump in and add to this, but I, I know that the Lazarus project in conjunction with, I think, Roger Easton, Alex, does that sound right, did some testing on the amount of damage that comes from the modern LED systems. And at most, it was the equivalent of being exposed to just like, fluorescent white light, like out in a room for 30 seconds, does that sound right, Alex? I know that some of the older preconceived notions of the potential for damage comes from the very, very, very first generation of multispectral imaging systems that did have bright white light that was then filtered out to get the different spectrum, rather than LEDs. And as with any digitization effort, if you’re shining bright white light continuously on an object, it produces heat. And I don’t have to tell you that that’s a bit of a bummer for a medieval manuscript. And so, but since that original generation, so much work has been done to avoid that particular type of damage.

Alex Zawacki 27:37
Yeah, the big threat with the first generation wasn’t even the the actual light itself damaging the manuscript, it was the heating of the manuscript that might result because they tended to use things like halogen bulbs, because they needed to get a broad spectrum light, because as Helen said, instead of imaging and discrete wavelengths, they would bombard it with broad spectrum light, and then filter out everything except one particular section of one particular wavelength. That means that that wavelength needs to be very bright, because it means you’re going to be dropping filters in front of it. And so that means that the light you’re using has to be extremely bright, which means that you’re going to be working with something like halogens, and you’re going to have long exposure time, long periods of exposure to that light, which can result in heating of the manuscript. Fortunately, we don’t have that anymore. LED technology, as Helen mentioned, means that we can just utilize the specific wavelength that we need rather than expose its broad spectrum light and filter everything out, which both limits the illumination that the manuscript is exposed to, but also because we’re using LEDs and that halogens means that the temperature of the manuscript isn’t changing significantly or measurably during the course of a multispectral imaging session, both of which means that it’s it’s safe.

[Music interlude]

Hannah Busch 28:53
Another question that I have, and of course, people might think about what they have in their collections, and that they want to use it. What are the limitations of multispectral imaging?

Katie Albers-Morris 29:03
One of the things that we’re actively working to fix is the need for you know, a very dark room and the affordability of the systems, right, they’re not they’re not small and they’re, they’re not cheap, and you can’t have any outside light. So certainly, that’s something that you know, the teams involved are working on trying to fix making it just a little more, I guess, user friendly and even you know, smaller archive friendly. Especially because if you’re an individual person who has one thing that you need done you’re don’t necessarily have a closet that you can set up a whole camera system in. But we do do a lot of work in closets.

Helen Davies 29:47
Yeah, as Katie is saying, there has been a lot of time spent in closets. You go to these beautiful locations and you’re like cool, I saw yet another closet, it’s great. Um, but the other so if we’re thinking about the difference between the bigger systems and the smaller systems, right, the smaller system, Tania has built a portable darkroom for it. So there is like an outside casing for it, that kind of just fits over the system. But what that does is it dramatically limits the size of the document that can be imaged. Or you end up kind of working very closely with a curator to be able to manipulate the document around the system, which doesn’t fit with the document and it it can be quite cumbersome. So, you can get the small and portable side of it, but then you lose the ability to work with larger documents, whereas the larger systems can obviously handle a much more wide variety of this. types of documents by as Katie says like require more specific conditions to set up.

Alex Zawacki 30:52
And there are particular substrates and forms of documents that MSI is much better suited to than others. It’s very, we’ve tried working with things like oil paint in the past, it’s really not very good for that it doesn’t see through oil painting very well at all. For that you’d really want to use something like, X-Ray Fluorescence or XR-D, it’s much better suited.Part of the reason that we tend to work on so many medieval manuscripts is because many of us are medievalists, but also because parchment, which is the standard writing substrate for the Western Middle Ages, is very good at holding traces of erased and damaged text whereas paper really is not. So even if you have something like the thick rag paper of the 18th and 19th centuries, which obviously holds impressions and erasures much better than say disposable A4 paper now, it’s still nothing like parchment. So we have been working on William Blake’s Four Zoas manuscript, which is in the care of the British Library, again with Christina Duffy, who is fantastic. And it’s a very challenging situation because the rag paper can be erased, it can be physically scratched off so that it can be rewritten. And it just doesn’t hold the data the way that parchment does, which even if something is very carefully, chemically and physically scraped off, that’s a very good that a medieval piece of parchment is still going to retain retrievable traces of what was there before, that’s not necessarily the case with with parchment, it’s much more challenging. So anything early modern to modern on paper is going to be a challenging situation for MSI. It really has its particular home in the Middle Ages. Of course, there are wide ranges of applications outside of it, but it does have particular substrates, and in particular, materials like not oil paints that it really prefers. And to which it’s best suited.

Helen Davies 32:47
To tie what Alex is saying back to your earlier question about working with documents at the University of Rochester. One of the genuine strangest moments is: I got called down to the English department office because a package had arrived. And I went and collected it and came back to the lab and opened it up. And it was a series of medieval map fragments. Well, it turned out to be medieval map facsimiles. It turns out, it took me like a week or two to track down what on earth these were, but that Chet Van Duzer had sent to the Lazarus project for the purpose of testing how well we could see through some of these paints that were used before deciding that we would try and image a particular map, which could not be removed from its backing. So yeah, yeah, sometimes it doesn’t work and sometimes you get little mini mysteries sent to you. Sorry, continue, Alex.

Alex Zawacki 33:49
Specifically, it was lead white pigment, which is why the facsimiles had to be made in Portugal and shipped to the US because there were various legal restrictions on playing with lead white pigment in the United States. But the original map has lead light on it and we wanted to try and see if we could see through it before we made the whole trip over to play with it in person. And so yeah, so it was lead white exploiting legal loopholes shipped to us from Europe.

Hannah Busch 34:16
That’s an amazing story. Um, I mean, you all three are from like, have ahumanities background. And you work during your PhD at the Lazarus project. So, I I would be really interested to hear how you incorporate MSI into your dissertations.

Helen Davies 34:39
So yeah, so, I think Alex and I took her fairly different approach for including multispectral imaging into our dissertation. And I know Katie is currently experimenting with the format of what her dissertation will look like. But for me, so my main research focuses on medieval mappa mundi. And the particular document which my dissertation is on is on the Vercelli Mappa mundi, which has been illegible since at least the 1970s. We do know that in the 1920s, there was an Egyptian Prince Youssouf Kamal, who took a series of photographs of the map at that time, and it still was legible. However, the map wasn’t rediscovered until 1908. So anyway, there was a brief period of time where this document was legible. And since that time period, nobody has been able to read it. So one of the things that I have done is I, the Lazarus project imaged the Vercelli map in 2013 and 2014. And then I, as part of my dissertation, processed the images and then used those images to form the large dataset, which I then used for my dissertation and kind of as a jumping off point. And then, just as a little side note, Dr. Heather Wacha. and I will be finally completing a digital edition of the Vercelli Mappa mundi for Digital Mappa hopefully within the next year, and that can then be freely accessible to anyone. However, the dissertation has three traditional kind of literary historical analysis, kind of tied into the document, but not based solely on that particular document that uses that document as a jumping off point for larger discussions of medieval cartography. And in one particular chapter, a very strange image of the King of France riding an ostrich on top of a mountain in North Africa. And then I have a further chapter that is just talking about the technical process of imaging the map, and recovering the map and then what I plan to do with it in the future. And then I had to create a rough draft of a digital edition that I hope to God nobody else ever sees, of the map as part of my dissertation. And so there was kind of, there was a traditional component of it, there was a digital write up, or there was a write up about the digital process, and then there was a digital project for it. Correct me if I’m wrong, Alex, but yours is going to resemble a much more traditional English department dissertation?

Vercelli Mappa mundi before
Vercelli Mappa mundi before MSI processing (preserved at Archivio Capitolare, Vercelli).
Vercelli Mappa mundi after MSI
Vercelli Mappa mundi after MSI processing (Lazarus project/Helen Davies).

Alex Zawacki 37:16
Yes, I’m actually primarily incorporating Lazarus into my dissertation by not doing so at all. For me, Lazarus, Lazarus work has been primarily book chapters and articles. But ultimately, I didn’t end up finding a way to work it into what I wanted my dissertation to be on. In part, because the pandemic, there were some manuscripts I would have liked to take a shot at imaging that just didn’t manifest and won’t have time to do before my dissertation has to be done. And so my dissertation, yes, is going to look much more like a standard traditional dissertation. Oh, yeah, I pass it to Katie.

Katie Albers-Morris 37:49
Yeah, I’m still, I haven’t turned in my prospectus or anything yet. So I’m still kind of early in the dissertation stages. But I am anticipating that mine will be much more technical, mostly because I’m pretty strongly considering all tech jobs are going into, you know, technical industry kind of things. So I’ve been thinking a lot about the uses of MSI, but also 3D to kind of create, like virtual reality experiences of archives and things like that. So that’s kind of the direction that my dissertation is going in at the moment. They may not, you know, and depth that way. But having the Lazarus project available to me means I have the opportunity to pursue that. And since I’m kind of doing it from the beginning, hopefully, it’ll work out as soon as I can travel and actually get my hands on some objects.

Hannah Busch 38:44
I would say, one last question. I know you’ve already mentioned a lot of very interesting projects. But if you can all like pick one favorite or most curious finding you had, so what you could reveal with multispectral imaging,

Helen Davies 38:58
I go first because I’m going to geek out about a very predictable topic and the Vercelli Mappa mundi is my baby, and I love it and it’s incredible, and which is like maybe a weird way to talk about medieval manuscripts. But I think it’s really really, really cool to be able to see a lost mappa mundi kind of emerge from, you know, a faded document, a damaged document. The first time I saw this map, I was given the RGB files for it and told to try and kind of read through and transcribe some of it. I was like, you can’t read anything. And then after some multispectral imaging, a lot of it pops out fairly quickly and fairly clearly, some of it takes a heck of a lot of work, but some of it kind of you can read fairly quickly. And it’s just so cool to see kind of the, you know, to use a slightly overly grandiose phrase, a bit like a, like, a mosaic of information and visual Encyclopedia of the medieval understanding of the world just appear before you. And it was just it’s very cool. And, and I’m sure Alex and Katie have radically different opinions about what is the best project they’ve worked on. But that’s mine.

Alex Zawacki 40:12
So I actually have to think because the first two that come to mind, I’m not allowed to tell anyone about. So I have to think about the coolest one that I’m allowed to talk about, which will take me a moment.

Katie Albers-Morris 40:17
Honestly, one of the coolest ones that I can think of is what Helen was talking about, like the Vercelli Mappa mundi. I did not expect to see, you know, a King riding an ostrich on a mountain in North Africa. And, you know, I was mostly living vicariously through her at that point in my education, but definitely, in my top like favorite things that have come out of nowhere.

Helen Davies 40:44
Actually, you know, to kind of jump on what Alex was saying earlier about the particular Smithsonian manuscript that was sent to us, it was actually really cool, because the imaging was so challenging. I mean, I can’t read any of the languages this manuscript is written in. But it was really cool to see the results of that particular imaging project. Because not all of the pages were visibly palimpsested, the imaging had been so challenging. And then Alex did the, Alex did all of the processing for that I was gonna say the bulk of the processing, but I think Alex did all of it. And he got incredible results. And so it was really cool to see this like before and after, where you’re like, looking at some pages, and you’re like, I don’t know, maybe there’s something there. And then Alex reveals an entire page full of text. And so it’s just like the most recent example of like, the stark before and after. And I think Alex did all of that processing in Hoku, as well. So it was kind of cool, not only to see Alex’s dramatic before and after, but also it was the first time I’d seen Hoku used to that kind of dramatic potential as well, which was awesome.

Alex Zawacki 41:51
Yeah, that was really my Crash Course and Hoku because there were so many leaves in that manuscript, that processing each one individually in ENVI would have been an impossibly long task. I mean, it was a very long task in Hoku, but an ENVI it would have taken ten times as long. So yeah, I processed the entire thing in Hoku. I did take some pages that I couldn’t get excellent results out of in Hoku and I reprocess them in ENVI, but 90% of the work was done in Hoku, which, like I said, really forced me to learn how to use Hoku very well. And yeah, that had some really dramatic results from the before and after from text that was barely visible or not at all visible in the original images to the under-text to being totally clear with the over-text vanishing completely was a very cool pop.

[Music fade in]

Hannah Busch 43:02
And with that we are the end of the episode, thanks for listening to Coding Codices a podcast by the digital medievalist postgrad committee. And thank you so much, Kate, Helen and Alex for joining us in this conversation about multispectral imaging and manuscript studies. If you want to know more, and if you’re looking for links to things mentioned in this episode, feel free to visit our website at podcast [dot] digitalmedievalist [dot] org where you can also find more episodes of Coding Codices. You can also get in touch with us by emailing to dmpostgrads [at] gmail [dot] com.

[Music fade out]

Credits

Music:

Graphics:

  • Featured image: Helen Davies and Alex Zawacki, ed. by Tessa Gengnagel.