So Yukinori Kawae is research
fellow at the Research Center for Cultural Heritage and Text,
Graduate School of Letters, Nagoya University, Japan. And he first saw the Giza
pyramids way back in 1992 as a 19-year-old traveling
from Japan to study the site. He moved to Cairo straight
out of high school. And since then, Yuki's
spent 16 years in Egypt and graduated from the
American University in Cairo with his degree in Egyptology. And that was under one
of our former speakers this semester, Dr. Salima Ikram. He started his academic
career as a member of the Ancient Egypt Research
Associates led by Mark Lehner, and he was one of
the supervisors of the Heit el-Ghurab
site, the so-called lost city of the pyramids just
to the south of Giza. Since the earliest stage
of the introduction of 3D technology into the
field of Egyptian archeology, he worked with 3D surveys of
ancient megalithic structures. In 2006, he joined an
interdisciplinary research project to produce 3D surveys of
the tomb of Queen Khentkawes-- that's that stumpy
pyramid of Giza, the so-called fourth pyramid-- and also in the worker's
cemetery of Giza. And in 2008, at the request
of the Supreme Council of Antiquities, as
it was known then-- now it is the Ministry of State
for Antiquities in Egypt-- he formed an industry,
academic, and government project and successfully completed
comprehensive 3D documentation of Egypt's oldest pyramid,
the step pyramid of Djoser at Saqqara, which is
now under restoration. In 2013, he expanded his
collaborative research to include a Japanese
TV production company. With a crew from
TV Man Union, he climbed the Great
Pyramid of Khufu to get data on its core masonry. Currently, he's working with the
Czech Institute of Egyptology, led by Dr. Miroslav Barta-- another name you
may find familiar because he was another one of
our speakers this semester. So I love the way
our speaker series is coming full circle this term. And he's leading an
interdisciplinary approach that incorporates
computer science and applied math to record
the pyramids at Abusir, the royal cemetery
from the fifth dynasty. It's about 11 kilometers
south of Giza. In 2016, Yuki was
selected for the singular honor of being a National
Geographic Emerging Explorer. In 2017, his team carried
out the first drone survey at Giza in collaboration
with TV Man Union and successfully produced
the first detailed 3D data of the Giza pyramids. A TV documentary
inspired a young Yuki to become an archaeologist, and
now he is the one on screen, appearing in Japanese
programs and giving talks about Egyptian
archeology and the history of ancient Egypt. And that's what he's here
to do for us tonight. Please welcome Yukinori Kawae. So thank you,
Peter, to introduce comprehensive my career. So hello everybody. I'm very pleased to
give a presentation here at Harvard University. Because, you know,
this university is a place where Professor
George Reisner was teaching. So he was one of the
master archaeologists who worked at Giza for a long
time and also established scientific and systematic
survey and the mapping in Egyptian archeology. So the general public,
and even Egyptologists, tended to attach
little importance to surveying and mapping
compared to discovery. But they are, of course, some
of the most important tasks in the field of archeology. The father of Egyptian
archaeologists, Sir Flinders Petrie, said that
most indispensable and important in the
tools she brought to an archaeological
site was a camera. This is because, unlike
line drawing and narrative descriptions, which are
archaeological interpretation, a photo can contain the variety
of the objective information. So recently, digital cameras
contain much more information than previous thought. So a photo contains
not only shape and the color
information, but also camera parameter and the
GPS information [INAUDIBLE].. So by using such information
with the structure from motion technique, which
I'm going to talk later, we can now produce 3D data. So furthermore, by attaching
the camera to a drone, so it is possible to obtain
image data over a place where access was
previously impossible. So we can now
obtain 3D data that has never been produced before. So this video footage,
you now think, is the top of the
Khafre Pyramid, second pyramid of Giza. So even though it is a
pyramid, everyone knows, no one was able to see the
detail of this place. This is actually the first
time to show this video footage in the US. And by using the image
photographed by drone, we successfully produced the
3D data of Khafre's Pyramid. So today, I will
introduce the history of our activities of 3D
surveys of Memphite pyramids with most recent, this 3D
drone survey of Giza pyramids. But first, I will talk
about our normal 3D survey using terrestrial time-of-flight
laser scanner at Giza. So in 2006, Ancient Egypt
Research Associates, led by Dr. Mark Lehner,
established Giza laser scanning survey team
with Japanese consulting to scan tomb of Queen
Khentkawes I at Giza. I was a member of this team. So my 3D documentation
of a Memphite royal tomb began with a 3D laser scanning
survey of tomb of Khentkawes. So this has happened. So Queen Khentkawes is
a mysterious figure, having ambiguous
title which, depending on the interpretation,
means either she was a mother of two kings
of Upper and Lower Egypt, or she was actually the King
of Upper and Lower Egypt and also mother of King
of Upper and Lower Egypt. So she may have
reigned over Egypt for a short time after Menkaure
and his successor, Shepseskaf, at the end of the
fourth dynasty. So our team, focused
on 3D scanning Khentkawes' giant
mastaba-like tomb, which is composed of two parts. Base cut from natural limestone
bedrock, this part, 45.5 by 45.8 meter,
and 10 meter high. And superstructure rising in
11 course of limestone blocks to a height of 7.5 meters. So her tomb stands at Giza. So southeast of Giza plateau
and near the central field of the Giza cemetery
and the central valley. So using three laser scanners,
our team work for two weeks to capture her
monument to produce the first detailed
archaeological data of this unusual tomb. Previously, Egyptian
archaeologist Selim Hassan excavated the base of the
monument in 1932 to 1933, and published 100 to 200
[? stigmatic ?] maps. So later, Vito Maragioglio
and Celeste Rinaldi, Italian architects, also studied
architecture of the tomb. So their plans on
the section are only the one today that shows
architectural detail and includes specific
measurements. So this is our updated
orthophotographic plan produced from our laser scanning
data which, as you can see-- so very different from
previous archaeological map. And it is interesting to note
that her tomb is not square, but somewhat
parallelogram in shape. The reason for
this shape may have been the result of
quarrying activity along the natural fissure
once running around her tomb. So this is an
elevation of Mokattam. Onto the south face
of a bedrock pedestal, a series of tall, broad panel. And the [INAUDIBLE] and
the recessed niches, we know it's a palace
facade motif were carved. So this motif commonly
appears on sarcophagi and also the mastaba
of the early dynasties. The panel facades, decoration,
and the most of which has been eroded here,
as you can see here. But the remains of the niches
and the panels identify that-- like here, at the
far upper east end, and also here at the very
bottom of the southern face-- so this tells us
that it was intended to create eight broad,
decorative facades with eight simple facades like that. So reconstruction
of palace facades indicated by white line,
while this red line indicates the remains of
the niches and panels. So our first 3D
survey, in 2006-- it's more than 10 years-- clearly indicated that
this survey technique is very useful in the field
of Egyptian archeology. But at the same time,
it became also clear that 3D data is
not easily handled without expensive software and
the professional expertise, due to the massive
amount of data. So we attempt to solve this
problem next project at Abusir. So in 2015, as a summer
project of Abusir archaeological
exploration conducted by Czech Institute
of Egyptology, we again, Japanese consulting,
initiated Abusir 3D survey for 3D documentation
of the site pyramid. The first season
of project focused on exterior of Neferirkare's
pyramid, which is the largest pyramid at Abusir. So Neferirkare was the third
king of the fifth dynasty. So his pyramid was
unique appearance. And it's thought to have
originally been constructed using step pyramid style. But it was later altered to true
pyramid, for unknown reason, with base lengths of
105 meters, and rising to a height of 72 meters. So Abusir is located about
11 kilometers south of Giza. It served as one of the
main area of cemeteries for ancient Egyptian
capital city of Memphis, especially the [? first few ?]
dynasty of the Old Kingdom. The Czech Institute
of Egyptology has been conducting excavation
at Abusir since the 1970s. The excavations are
currently directed by Professor Miroslav Barta,
who gave a lecture here last November. I'm sure some of you
actually attend the lecture. So academic research
using 3D technology has become popular in the
field of Egyptian archeology. But also, this new
technology allowed for more detailed documentation
than conventional hand mapping, like the one I showed you. It has yet to be
fully integrated into archaeological research. There appear to be several
reasons for this situation. One of the crucial
reasons, I think, is that the 3D surveys
normally have no other option than to scan a subject
as thoroughly as possible in the hope of
gathering whatever data might be considered relevant
to [? projectable. ?] But this results in a large and
unnecessary volume of 3D data. So that is not easily handled
without expensive software and the professional expertise. So in essence, we can call
this overscanning of subject. In offering a solution
to this problem, we promote interactive
interdisciplinary research program conducted by
archaeologists and engineers, computer scientists, as
well as software engineers and applied mathematicians
in order to properly document a monument. Actually, my team-- I'm the only archaeologist,
and the other team members are computer scientists
and applied mathematicians and physics and-- it's not people from archeology. So first of all, we
checked the resolution of our previous 3D data. This is also a
photographic image of 3D data of Tomb of
Khentkawes at Giza. The number of the
point of the data is 250 million point cloud. So this what we call
[INAUDIBLE] point cloud, because it is a set
of the data of the point in coordinate system. In three-dimensional
coordinate system, these point are usually defined
by x and y and z coordinate. So this is an enlarged
image of eastern side of tomb of Khentkawes. The resolution of the
area in the shadow is 6.0 meter pitch
per centimeter, while it is 3.7 milli
for the area in the sun. So in the Abusir 3D survey
project, we attempt to apply-- quantitatively
model the strategy instead of relying only on
experience of 3D surveyors in order to avoid
the overscanning. So we call this strategy
the optical scan plan. So I was asked my team
member, when we decided to carry out Abusir 3D survey-- so they asked me, "So
what is your plan?" And then I explained
my plan is, you know, [Egyptian for blah blah blah]. But my team member said,
"No, this is not a strategy. This is just grit and guts. You have to think
about strategically." So he explained, and my
team member explained, optical scan plan. So what is it? So optical scan plan, however,
3D information of the subject is needed beforehand,
though we are facing with a kind of dilemma,
that a 3D model is required in order to appropriate
the plan of 3D scanning survey with
mathematical programming. So then, we choose structure
from motion technique, which is a technique to produce
a 3D model from 2D image-- I mean, digital photo-- easily. So we produced large 3D
model with this technique. On the left here
is a process flow for 3D modeling of
preliminary survey. We first take the
photo, and the two collected in the
photo collection. Then we use the structure
from motion technique to produce 3D point cloud data. And then, using multi-view
stereo, we use the mesh model. After that, we do down sampling. And then we made in
the lab 3D mesh model. So this is a flowchart for 3D
modeling of preliminary survey. And on the right,
this is scan planning with mathematical
programming framework. So this is our first step. We visit the site, obviously,
to quickly photograph the pyramids. We spent only a few hours-- I think two hours
or three hours-- to obtain digital image and
video footage of the pyramid, Neferirkare pyramid,
using digital camera. This is my colleague
Dr. Yoshimura and he's an engineer
from Kansai University. And he used his own digital
camera to take the photos. So the data, then,
analyzed using [INAUDIBLE] and multi-view stereo software. And then we produced 1.69
million point cloud dataset from 624 digital images for
our mathematical optimization process. After that, we choose
regions of interest-- ROI, what we call it-- in which core masonry
still remained. The red colors in
this image indicate our archaeological
region of interest. And I would say this
process is very important. Indispensable process. Otherwise, 3D surveyors often
scan a subject as thoroughly as possible, as I said previously. So next, we downsize obtained
in the mesh of the target area and made simplified model
from around 5,000 triangulates for mathematical optimization
process, as shown here. And here is Dr. Dan, my team
member from Kansai University. He's an applied
mathematician who planned the mathematical
optimization process, optical scan plan. And here, he's explaining
to me the process flow for 3D modeling of
preliminary survey. So our first priority is
to scan entire pyramids as much as possible, but
not overscanning, as I said. And next, we measured
region of interest with a certain density
of point cloud. In consideration of
these conditions, Dan mathematically obtained
appropriate scanning point. So first, we identified in the
camera position of 624 photos we took. These camera positions were
used as the [INAUDIBLE] of scanner position. So then, Dan mathematically
solved the optimization problem. So eventually, it became clear
that we needed only seven scans to measure all face of
Neferirkare pyramid. The image here is optical layout
of seven scanner positions. So white dot-- this one-- our [INAUDIBLE] scanner
of 624 positions. And the red dot, like here,
depict scanner positions. So we also used
visualized effectiveness of our scanning plan. So brighter color, here, were
receiving the higher number of scanning laser beam
from these seven positions. So we did the scanning
plan, so we finish it. So then we are ready
to go to Egypt. But we had a problem. Our scanner looks funny, I know. Our scanner, so [INAUDIBLE] the
photo [INAUDIBLE] laser scanner was too heavy to be
check-in baggage, as it went over 40 kilogram. So the baggage allowance on
all the airlines to Egypt does not exceed 30 kilo. So what we have to do is, we had
to purchase a seat for scanner on board. So flight attendants ask me to
fasten seat belt for scanner. Then it was too bulky. And then I needed
to get an extra belt and finally fasten the seatbelt. So finally, we get it to Egypt. And then, my team
member Dr. [INAUDIBLE] and his graduate
student, [INAUDIBLE] began to set up the scanner. In this optical
scan plan, we also emphasize human interaction
to give feedback from the site for
flexibility and effectivity, changing the plan
according to needs on site. This is called Human-in-the-loop
in the machine learning, which is defined as a model
that requires human interaction. But we call this
archaeologist-in-the-loop. So, for example, when we realize
that our scanner could not be put in the
position we planned, or when we thought
we should scan the pyramid from
extra positions, so we homed it to Dan from
the site, normally by Twitter. And then, he'd calculate
it and then model it with optical scan plan. So this is
archaeologist-in-the-loop. So our first priority was
to scan entire pyramid. But however, from the
beginning, optical scan plan indicated that some area, such
as the top of the pyramid, can be scanned from the
ground-set laser scanners. So if we use a
drone, of course, it would be very easy to obtain
the image data for structure from motion for the
production of the 3D data. But we are unable to
use one that time. So instead of a drone, we used
small, 180-degree cameras which we attached to a fishing rod. So a fishing rod in the desert,
I know, it looks very weird. But it's actually effective. It's very effective. Yeah, you know, my
Egyptian colleague said, "What are you doing with a
fishing rod in the desert?" And of course,
they are wondering, these guys are crazy, or nuts. But anyway, it was effective
for obtaining the image data for the top of the pyramid. So this is, finally,
we produce 3D model of Neferirkare's pyramid. The average height
of the pyramid in its current condition is
approximately 41.5 meter. And then average base
length of the pyramid in its current condition
is approximately 92 meter. So original height is
considered to be 72 meter, and its base is 105 meter. So it is a bit hard to believe
a 30-meter high structure was weathered and lost. So I think we probably have to
reconsider its original height and the shape. And regarding the
problem of overscanning, the number of the point
is only 47 million. This is a very small
number in comparison with tomb of
Khentkawes, which is much smaller than this pyramid. So Khentkawes' tomb
is only 80 meter high, and its base is
about 42 by 44 wide. But the number of the
point cloud is 250 million. And regarding archaeological
region of interest-- so like here. This is an enlarged
image of the south face of Neferirkare's
pyramid, which is one of our archaeological
region of interest. So this image clearly shows that
the structure of the pyramid is well defined. So according to the US General
Service Administration-- GSA-- Building
Information Modeling-- BIM-- Guide Series
03, 3D Laser Scanning, the general quality level of the
scanned 3D point data should be described based on
point of density. So this table shows a
project definition matrix that can be used to identify
how 3D data can be used to support project objectives. According to this
matrix, the scanned the data can be
in two full levels based on the density
of the point crowd. And our point crowd is between
level two and level three. So our archaeological
aim was successfully achieved, our project. So we referred to this General
Service Administration Building Information Modeling to evaluate
our archaeological 3D data. But I think that
we should actually establish Archaeological
Information Modeling, AIM, Guide Series, by ourself. So this will be our
next research subject. So this is a 3D model of
Neferirkare Pyramid which was created by one of
our team members, Mr. [Ishikawa, CG artist. We normally produced
two types of data. One is for academic
research such as orthophotographic elevation and
the plan, which I showed you. The other is for public
lecture, like this CG. So we normally say
that this is CG image. And so when the 3D survey
of megalithic structures like pyramids is conducted
using terrestrial time-of-flight laser scanner, it is not easy
to capture entire structures with a monument. Therefore, in our
project, we applied mathematical
optimization approach for 3D recording of the pyramid
in a cost-effective way. The usefulness of this
newly developed plan become very clear in the
archaeological survey. So we continued. In 2013, we also started to
collaborate with Japanese TV production company,
TV Man Union, to establish Giza 3D survey to
obtain the 3D data of the Giza Pyramid. As many of you know, there
are numerous hypotheses on the construction
of the Great Pyramid. But only a few
actual surveys have been conducted, such as
Egyptologist Flinders Petrie in 1880s, and Vito
Maragioglio and Celeste Rinaldi in 1960s, and Gilles
Dormion and Patrice Gordon in 1980s-- but is published in 2004. So moreover, these
archaeological surveys have only focus on external
part of the pyramid and the inner structures such as
the chambers and the corridors and the passage. But no observation of the
pyramid has been made. The masonry of the
pyramid of this period, the fourth dynasty
of the Old Kingdom, has not yet been fully
studied, because most of the pyramids in this
period are well-preserved. So you can see the
section of the pyramid. But general opinion first. So core masonry may consist
of a horizontal laid block. And second-- so pyramids
may have a core step. We know the pyramid is a core
step from the queen's pyramid. And also, third, the core may be
built with a [INAUDIBLE] layer. The importance of the study
of the inside of the structure is that core masonry is closely
related to the construction methods, namely, the type of
the ramp Ancient Egyptians would have employed. For example, straight
ramp is probably suitable for setting
horizontal reversing block. Or debursting ramp,
or zigzag ramp, can be easy to lay stones
that are core step. Or wide acceleration
layer core would be appropriate for
overlap spiral ramp. So each theory has, of course,
disadvantages and advantages. But certainly, without the
study of the core masonry on these early on paper. There is a place, however,
where we can actually observe the core masonry
of the Great Pyramid. But if you want to
observe this, you have to climb up the
pyramid, because it is located at northeast
corner of the pyramid, and about 80 meter
from the ground. In 2013 and the 2015,
Japanese TV Man Union, TV production company,
had the opportunity to climb the pyramid. And I was with them. And they climbed up
the pyramid to obtain the data of the masonry
structure of the pyramid. So this the video footage
our cameraman filmed. I'm often asked how long it
took to climb the Great Pyramid. It took five and 1/2 hours
to climb up and descend. Because we measured every
course of the pyramid to check the height
of the stone, which had previously
been measured only by Flinders Petrie
and George Goyon. It was quite the
experience, because it is an ancient monument. We had to climb without
pitons and safety rope. And the slope of
the Great Pyramid is 51 degree, 50 minutes. But as you can see here-- when you see the slope
from the top to down, yeah, it is almost vertical. I was really worried
about the cameraman might fall from above, because
he actually filmed like that. And then even he-- before he climbed the
pyramid, he confessed to me, "I'm terrified of heights." But he also said, "But
if I have a camera, there should be no problem." Yeah. In fact, he managed to
film without any problems. So [INAUDIBLE]. So we arrived at our target
area, a small half-open space called notch. The notch is located over the
104th course of the northeast corner of the pyramid. It is often said that stones of
the pyramid were perfectly set. Yeah, I think that some of
you have visited in Egypt. Then tour guides actually
mention to you between the stone, even the knife cannot
actually-- the putting it. But as you can see here,
the stone inside pyramid are not perfectly
set and aligned. The masonry is loose. And first of all,
interestingly, the notch has a crevice in
the west that led to another open space called
the cave, inside the pyramid. So these places can show
core masonry structure. So here's the cave. Again, as you can
see here, stones that are inside
the pyramid, they are not perfectly set on the
lines, even not oriented. So these areas are
important in study. Because, as I mentioned,
previous archaeological surveys of the Great Pyramid
have only focused on external part of the
monument and in space such as the chamber, and the
passage, and the corridor. But no observation of the
core pyramid has been made. Therefore, if we can produce
3D data of this area, this will be the
first data produced of actual state of the
core of the Great Pyramid. But since it was not
pure archaeological-- the survey, I didn't bring any
survey equipment at that time. But I really wanted
to record this area. So I asked my team
member computer scientist whether it is
possible to produce 3D data from this video footage. My colleagues, Dr. [INAUDIBLE]
and Dr. [INAUDIBLE] said it is possible. And also, TV Man
Union kindly allowed us to use this video footage
for academic research. So 20 minutes video footage was
split into 30,000 image frames, out of which we
selected 300 image shots using several smooth
motion of the camera. And then, we used these for
structure from motion technique to reconstruct 3D model
of this study area. But Dr. Yoshimura kept
complaining one thing when he produced 3D data. So he said, "You
are bothering me. Because in every
frame, I see you. You are there." But what can I do? This is not my intention. This is just a TV documentary. But anyway, so finally,
we could create 3D data with this structure. This is our 3D data
of the notch and cave. This is, of course,
not imaginary CG. It is not. But it is purely based on
archaeological structure of the Great Pyramid. As this shows,
stone in this area are much looser and irregular
than previously thought. And as I said, even
they are not oriented. As I mentioned before,
general opinion regarding the core masonry
layer, the first core may consist of
horizontal length block. The second pyramid
may have a core step. Third, core may be built
with acceleration layer. However,
orthophotographic section we produced from 3D data
show a different structure from previous theories. Our interpretation
is that this cavity would have been used for
the chamber method, which have remarkably accelerate
the work in an economical way. The possibility
of the application of this construction
method has already been suggested by Dr. Miroslav
Verner, ex-director of Czech Institute of Egyptology. So our 3D data and
archaeological interpretation would support his idea. So the chamber method was
actually well-known technique throughout Egyptian history. In the pyramid of
Middle Kingdom-- pyramid of Senusret I at
Lisht, irregular pattern of the masonry floor surrounded
by [INAUDIBLE] masonry is already known. So notch and the
cave seems to be similar to this construction. So even in the Old
Kingdom, we can see some example of this
method, such as the ambulatory of the Sun Temple of
Nyuserre at Abu Ghurob. So besides the
notch and the cave here, there are other
regions of interest such as upper part of
the Khafre's pyramid, where we can observe
the relationship among casing stone, backing
stone, and also core masonry. And the top of the Great
Pyramid also very important, because where we can
observe the inner structure at the top cross-section view. So this is a top. The Great Pyramid was
presumably constructed from 210 limestone courses. However, due to the loss about
casing stone and capping stone, the top of the
pyramid is currently approximately 12 meter
exposure of 200 first courses. And again, I photograph
the top of the pyramid. I took many photos. Then, we used these digital
photos, and then also video footage, for 3D
reconstruction using structure from motion and multi-view
stereo technique. But unfortunately,
this didn't work well. The image shown on the screen
behind me, as you can see here, there is a missing
part in the 3D data. For the production of the
perfect 3D data of the top, it is necessary to
photograph using a drone from more distance place
and from more angles. Initially, we are thinking about
bringing a drone into Egypt. But it was difficult,
because in Arabic, a drone is called a spy plane. So the Egyptian
authorities are concerned about the risk of a drone being
used for military purposes. So it was impossible
to bring it into Egypt from a foreign
country at that time. However, by a series
of happy accidents, we finally carry out
drone survey last year. First of all, I was,
as Peter mentioned, elected as Emerging Explorer
of the National Geographic Society. This is not my
personal achievement, of course, but achievement
of all my team members. So next, 13th anniversary
of the TV program-- Japanese TV program-- World Mystery
Discovery, [JAPANESE],, produced by TBS and
the TV Man Union. So TV Man Union
decided to collaborate with National Geographic
Society and support National Geo Explorer. So they supported
our project again. And we couldn't bring
a drone into Egypt, but TV MAN UNION found
a local drone company who obtained a special
permission from the military to use a drone in the country. So beside photographing
with drone, we had to set up and
obtain difference point for global navigation
satellite system, GNSS, which is well known as GPS,
Global Positioning System, but for integration
with 3D data. This is because if 3D data
generated from photograph isn't integrated with survey
data, it is just a 3D model, and its size and
orientation are unknown. So reference point
for drone survey were pasted on the pyramid
with duct tape, of course. They are easy to peel off, and
then no damage to the monument occurred. So these reference
points were then measured with total station,
then integrated with GNSS data. For each pyramid, we took
pictures in detail each place. So small-scale, you can see here
marks in these slides, which is the east side of the Pyramid
of Khafre, the position where we photographed with a drone. So for the pyramid of Khufu, we
shoot more than 7,000 photos. For Khafre's pyramid and
the Menkaure pyramid, we shoot more than 4,000 photos. So this is 3D image created,
again, by our team member Mr. [? Ishikawa, ?] CG artist. And again, this is
not imaginary CG, but it is based on 3D model
integrated with GNSS data. So each stone and each-- the stone will present the
original shape and the size. And then this is
our previous notch and the cave data
we also integrated within the drone data. So this is remarkable result. But how we can use this data
for archaeological purpose? Previously, only the sketch by
Vito Maragioglio and Celeste Rinaldi, Italian architect,
exist of the top. So I should emphasize
that this is the only place we can observe
the top cross-sectional view of the inner structure
of the Great Pyramid. But as you can see here,
it is very difficult to understand the
structure from this sketch. So we used digital photos taken
by drone for 3D reconstruction. So this is our
first detailed plan of the top of the Great
Pyramid from our 3D data. This plan is not just
point cloud image, but shows [? duration ?]
structures. This plan is produced
by [? Pict, ?] which is our newly developed
image processing technology. So [? Pict ?] can also
show color distance. The color of the point
expresses quantified the distance between the
point and an arbitrary plane. So warmer color, like
this brown and the yellow, means shorter distance from
the reader's viewpoint. And the cooler color, like
in the blue and green, means longer distance. From this [? Pict ?] image,
we made in line drawing and then numbered
every stone blocks. We counted in the 104
blocks from one to 86 on 200 first course, and from
87 to 104 on 200 second course. So by dimension of the area
is approximately 11.7 meter. And the 11.9 meter, this
is the present state of the top of the Great Pyramid. So our next question
is to find out how large this
area originally was for understanding relationship
among the casing stone and backing stone
and core masonry. So from the [? Pict-- ?]
sorry, this one-- we use, actually, the
similarity in geometry. So I'm archaeologist. I have no idea how to get
these geometry things. But as I said, I've been working
with a computer scientist. And then I gave kind
of a quiz to them. So then they just
answered within 15 minutes by Skype and email. So this is their answer. So we know original height of
the pyramid, which is 146.58. And we also knew
original lengths of the base, which
were recently being measured by Glen Dash from
Ancient Egypt Research Associate. And it is 230.392
meter on average. So this is a
mathematical formula. And the answer is 13.64. So this white area is the
original dimension of the top. We also calculated
the total weight of existing stone
and original blocks. The total weight of
the existing blocks is about 188 tons and if
density of the limestone is 2.6. And the total weight
of the original blocks was about 273 tons. In other words, the
[INAUDIBLE] rate was 68.95%. And about 31% of the
blocks are missing. But I should
emphasize that there's uncertainty in this value. So next, in order to understand
the structure of the top, we made a cross section of
every line of the blocks. There are cross section
landing east-west, like that. And then most of the
section clearly shows that the top of the
pyramid is not leveled. But there are two areas
that are very level, which are north end and
south ends of the top. And I carried out
the same procedure of the north-south
cross section again. And most of the section showed
in the top of the pyramid is bumpy. But there's one area
is very leveled, which are east end and west end. And east end is-- it's also leveled. But due to the blocks
of the 202nd course on the top of this area,
you can actually not observe the complete
cross-section in this area. So these area are backing stone. So backing stone are
generally considered to be frame that
support casing stone. So another interesting
point is that the center of the 201st course is recessed. And conversely, the
large stones were put in like here, the center
of the 202nd course, to create convex surface. So this is enhanced the image. But similar structure. Yet one that is
more sophisticated can be seen at the top of
the pyramid of the Khafre. You might remember the first
video footage I showed you. So this structure is
applied to improve, I think, in the stability,
which is probably not only for the top of
the pyramid, but also the main body of the pyramid. And as I said the
beginning of this lecture, there are many hypotheses
concerning the construction of the pyramids. However, only a few actual
surveys have been conducted. And also, previous study of
the construction pyramids tend to talk about the
whole construction at once. But we would like to demonstrate
construction technique of particular area such
as the top of the pyramid, because it is a 3D-- the data, we can actually
separate each course and then analyze quantitatively,
the course by course. And in the near
future, we would also like to publish this data
to promote further research on the pyramids. So before I finish
my lecture, I would like to give a special
thanks to Ministry of the State of
Antiquities and the Supreme Council of Antiquities,
and especially my Egyptian colleagues
for their cooperation. So thank you very much.
