Saturday, 6 February 2016

Digital archaeological drawing on the field with QGIS

I should have written this post since long, but time is always missing... The topic regards the digital archaeological (vector) drawing on the field. 

During the CAA conference of 2015, held in Siena Italy), I participate, among others, in the session 9A (Towards a Theory of Practice in Applied Digital Field Methods), moderated by +nicolò dell'unto (Lund University) and James Stuart Taylor. After my speech I was asked if we (Arc-Team), as a professional archaeological society, were really able to perform the digital documentation in real-time during an ordinary excavation. I answered that, at least in Italy, this point is very important for a professional society and that in normal conditions (but this can also happens during most of the emergency excavations) we complete the digital archaeological documentation directly on the field. The reason is simple and it is every year more evident: money are always less and less, as the time goes by, for cultural heritage matters; at least this is the trend of the last decade. For this reason, if on the one hand we have to try to counter this phenomenon, on the other we have to adapt our methodology to the current reality and this means to use the economic resources for the excavation also to produce the related documentation (without counting on a post-excavation budget).
The old video below (2014) shows how we manage the digital archaeological drawing on the field with QGIS.



The vector layers can be related to georeferenced photomosaic (bidimensional photomapping), or to georeferenced orthophoto (coming from 3D operations based on SfM/MVSR techniques). Of course orthophoto are the best solution, but currently the 3D work-flow with standard hardware is pretty slow. This is the reason why for almost all the palimpsestic documentation we still work with both the system: 2D photomapping and 3D SfM; depending on the time-table we have on the field, we choose the post-processing operations.
Within QGIS it is possible to draw vector layers in different ways. Basing on our experience (as you can see in the video), the best two best solution are:

1. to use the plugin Freehand Editing (if you want to experience something that is really similar to the old-traditional methodology, with the pencil and the paper)

2. to use the standard vector drawing tools (if you want to avoid too complex sahpes, like polygons with too many nodes)

IMHO, Freehand Editing plugin is a perfect solution for field operation, so that I am planning to add it into qgis-archeos-plugin, for ArcheOS Hypatia.

Monday, 1 February 2016

Arc-Team 2015 Iran-Mission: The Video

In June 2015 we had the pleasure to prove our abilities for the first time in the Middle East:

On behalf of the University of Innsbruck - Department of Near Eastern Archaeology, in collaboration with the University of Sistan and Baluchestan and at Iran's Research Institute for Cultural Heritage and Tourisms invitation we've documented one of the most important sites of Persian heardland:


The Palace of Ardashir Pāpakan (in Persian: دژ اردشير پاپکان‎‎ Dezh-e Ardashir Pāpakān), also known as the Atash-kadeh .آتشکده

Ardashir, also known as "the Unifier" (180–242 AD), was the founder of the Sasanian Empire. He was the ruler of Estakhr since 206, subsequently Pars Province since 222, and finally "King of Kings of Sasanian Empire" in 224 with the overthrow of the Parthian Empire, ruling the Sasanian Empire until his death in 242. The dynasty ruled for four centuries, until it was overthrown by the Rashidun Caliphate in 651. (Wikipedia)

The building is located two kilometers (1.2 miles) north of the ancient city of Gor.

The palace complex includes a pond, fed by a natural spring.

The structures dimensions are 116 m by 54 m. The three domes are almost 18 m high, the southeastern one is partially collapsed. The structure was built of local rocks and mortar with plasterwork on the insides. 

After positioning the site with DGPS we've documented the building room by room, applying the structure from motion technology, powered by free and open source applications.


In ten days we took about 30.000 pictures from the ground and from our drone, covering the whole building and two Sasanid Rock Reliefs: The Investiture Relief of Ardashir I. and the Equestrian Relief, showing Ardashir's fight against the Parthian king Artabanus V in 224.

Our three minutes long clip sketches the 2015-Mission of "Digital Archaeological Documantation of Iranian Monuments" during all it's phases.

We would particularly like to thank:
Sandra Heinsch-Kuntner
Rouhollah Shirazi
Walter Kuntner
Sasan Darvish Zadeh

Friday, 29 January 2016

Blender 3D and Cork, a duo that saves lives and rebuilds heritage


Statue of Saint Martin de Porres ready to be printed in 3D
I've always loved to study, no matter the subject. I constantly "lost" a lot of time of my life testing a technology or discovering a new way to do what I was already able to do in a conventional way. The fun of all this was to learn it and not necessarily to apply it. I was often asked why I was doing this and found myself in serious trouble to answer, which was something like, "Look, I don't quite know what this is, but I imagine one day I may need it”.

Some pieces before the boolean operations with Cork - base body modeled by Daniela Ayma
Time passed, and as I assumed myself as a nerd at 16 years-old, I did a lot of that, I studied things without knowing for sure what for I'd use them, but kept studying. It's been already 17 years doing that same thing,which is more than half of my life. In the meantime I realized that sooner or later an application might appear and it's best to study it hard and not to know where to use it than not studying hoping that inspiration will come from heaven.

Example of a boolean operation performed with high resolution meshes
The fact is that after a while the question "what is it for?" started to bother me deeply. It is expected that sooner or later people, bearing the intelligence that is inherent in them, will know where to use that load of knowledge.

3D printing performed with two materials (rigid and flexible). Thanks to Claudio Sampaio Marques (Patola)

If on the one hand I was becoming grumpy, impatient to explain others what a knowledge is for, on the other I realized that it's essential to have motivation to "sell" ideas, i.e., it must be very clear in our minds we intend to do when we call in other people in our personal projects. This requires knowing quite where we want to arrive.

Fortunately, the grumpiness disappeared and here I am sharing with you one of these projects. A study that began almost by pure curiosity and became a series of actions that have even saved lives.


The Problem of Boolean calculations in 3D

In Blender there is a so-called Boolean modifier. With it you can do calculations as "difference," "union" and intersection "using two 3D meshes. Who works with modeling and animation knows the term and also knows that if the meshes start to stay heavy and complex, the calculations usually have problems. Not unwillingness of developers ... they strive and ask for users to submit the "bug reports", but it's a little tricky to do for those who do not understand programming, or expects the program has to resolve this type of situation .


Globo Natureza: Prótese Animal
[VÍDEO] Nosso trabalho de criação de próteses para animais é tema do GLOBO NATUREZA!Link original: http://goo.gl/h3ykF7
Publicado por Cícero Moraes em Segunda, 4 de janeiro de 2016
Video featured on Globo TV's show Globo Nature where the case of the tortoise Fred is highlight

In my case Blender's Boolean had always functioned well. All I needed so far had been solved with the modifier and its basic operations. The problem began to appear when I began my studies on prosthetics. The first fire test I did was to create a hypothetical prosthesis for a digital skull that had suffered considerable loss of bone tissue. To me it seemed obvious that all would work well, but as I started working the Boolean on the complex mesh of a skull, I realized what a problem I had gotten myself into.

The Boolean and orthognathic surgery

In August 2014 I was contacted via email by Dr. Everton da Rosa, a renowned orthognathic surgeon seeking partnership for the development of a surgical planning technology using free and open-source software.

Initially he hired me to teach Blender, InVesalius, MeshLab and other software, all intended for the simulation of virtual surgery. Things went very well, until we have to simulate the cutting of bones. When the need to use Boolean appeared... crash! The tool bugged already on the simplest cuts.

Virtual ostheotomy performed by Cork
How could we solve this? I was a little embarrassed to see Blender not work properly, but the fact is that it was not originally designed for that type of work. We were talking about meshes easily reached up to 400 MB, how to proceed with the calculations of such an “elephant” like that, when the program was developed to work with a “mouse”? That's when I used a knowledge which until then was not, shall we say, very useful in my teaching activities of computer graphics... it was the knowledge related to compiling programs.

Out of pure intuition, I looked up if there was any standalone Boolean software. I found a topic in CloudCompare forum which spoke about Cork. Long story short, since I wrote a post about it for a while, I managed to compile the program and proceed with the complex meshes of calculations in Blender.

The only problem is that for Cork to work we had to export the meshes to use Cork by command line, or in CloudCompare's GUI, where it had been adapted. It was hard to teach and it gave more work for Dr. Everton, which had to forcelfully learn how to use the command line. And not just him, another health-area student of mine, Dr. Rodrigo Dornelles, Plastic Surgeon, also had to meet the pleasures of the black small screen.

Creating an Addon - who knows it, knows it.


We had solved this problem. We knew how to do very complex calculations, but we spent a lot of time doing them. That's when I thought about contacting my friend Dalai Felinto, an expert in Blender Script.

I explained to him how we did the Boolean calculation process, sent the codes of Cork and he began to study a way to carry everything into Blender. I thought it would take a long time, but he surprised us and gave us a test tool in a few days... who knows it, knows it.

Addon interface developed by Dalai Felinto (beta)

To our joy everything worked well on several different operating systems (Windows, Mac and Linux) and start to use Cork directly within Blender.

With a clear head I could devote myself to other studies and one was precisely the making prosthesis for animals, a fantastic project we were doing together with a number of professionals on the field of veterinary medicine and dentistry.

Obs .: I would like to thank my fellow Paulo Amorim, from CTI Renato Archer for compiling Cork for WIndows. I have no idea how to do this on this operating system.

Saving Lives with Blender and Cork


I never said it openly, but I used Cork addon in every project we did to create prosthesis for animals. Our team decided to keep everything secret until we are sure that the tool was good even. Fortunately it proved itself successful after many favorable results.

Stages of making prosthesis using the animal's beak as a volumetric subtraction parameter

The first animal we helped was Fred the tortoise. The shell we created for her was made possible thanks to the use of Cork within Blender. This enabled us to achieve a structure in record time.

The second animal was Zeca the stressed toucan that broke one beak after another. The third was the toucan Bicolino... then the goose Victoria and finally Verdin the parrot. So far it's been five animals and right now I'm a flight to Brasilia, where I will meet the team for a few beers and talk about the sixth and the seventh animal that we will save.

Here's two posts on the subject:



Saving the Heritage

The power of the tool goes further. We initiated a project with the Religious Art Museum of Santos attempting to restore missing pieces of religious artistic pieces that have been deteriorating over hundreds of years. We intend to rebuild dozens of pieces and at the end share with the community the technical and artistic results of the project through articles and publications.

Cork used to recovery historical pieces, such as missing toes of Christ present in the Museum of Sacred Art of Santos (SP)

Note: We thank the architect Fernando Gregório, Secretary of the Religious Art Museum of Santos for the welcome and trust. Personally I also thank Dr. Paulo Miamoto for his motivated spirit engaged in our projects, not just providing his vast scientific knowledge, but also material resources, to print the pieces that will recover the history of Santos and Brazil.

Obs.: Dr. Paulo Miamoto also translated this post... thank you very much!

Printing Cultural Projects

Another area that was highlighted by Cork Addon was the cultural project of the Facial Reconstruction of Catholic Saints. Print files of the busts have been prepared from Boolean calculations, featuring speed and accuracy in the process of creation of life-size busts.

Saint Paulina. Stages of making a life-size bust: 1) 3D Modeling, 2) Mesh preparation for printing, 3) Paint on 3D printing. 4) Presentation of the bust to the faithful. 3D printing: CTI Renato Archer, painting by Mari Bueno

Conclusion

Cork adapted to Blender opened up a huge range of possibilities, especially when we talk about 3D printing field. On the one hand the Blender native boolean tool is excellent for simpler parts and asking for a test results in real time, on the other Cork is powerful with more complex meshes and requires a great processing power.

Following the natural path of these tools, Dr. Everton Rose decided to share the addon, especially because we are developing a plan of courses aimed at application of computer graphics in healthcare. It is possible that soon the addon will available on a site like BlenderMarket at a very affordable price, so that the aid can cover the investment made to carry out the tool and who knows, may foster the development of others that come around .

The fact is that, in the end everyone ended up winning, believers, historians, artists, and most importantly, the animals had a second chance to regain life.

Tuesday, 26 January 2016

gvSIG 2.2 - 3DView plugin

Hi all,
i was playng around the new 3DView plugin of gvSIG 2.2. In this first version it is possible to translate your 2D map into a 3D viewer based on NASA World Wind SDK. Remember that the 2D map must be in EPSG: 4326 . The video below show how to import a vector layer.



The viewer is really fast and simple to use. The active-syncro between 2D and 3D world is a good choice. Since this first version the user is allowed to import vector data, raster data and 3D DTM (as raster). The developers want to improve the import function: features as anaglyph, voxel models, 3D symbols… will be added in the next release.

Thursday, 14 January 2016

ArcheFOSS VIII proceedings finally online!

Hi all,
Thanks to the effort of +Filippo Stanco and Giovanni Gallo, are now finally online (as Open Access) the proceedings of  the eight ArcheoFOSS, which took place in Catania (at the Department of Mathematics and Computer Science of the University) in 2013 (here the ATOR related post).
The event was organized by the Image Processing Laboratory and the proceedings are published by ArchaeoPress, as open access (here the direct link).
Here below you can see the front cover:

The front cover of the publication

In this publication you can find some articles of ours and in particular the analysis of the Taung Project under the aspect of the Open Research (here the slide of the presentation) and +Rupert Gietl's report about the 3D documentation of archaeological underground environment, related with WW1.

Have a nice day!

Saturday, 26 December 2015

ArcheOS Hypatia 3DHOP package (call for testing)

This short post is related with the previous one (about Nexus) and it is a call for testing for the deb binary package (arm64) of 3DHOP, the software developed by the Visual Computing Lab (CNR-ISTI, Pisa - Italy) to create interactive and multi-resolution web-galleries of 3D objects.
Here below you can see just an example of a gallery related with conflict archeology (WW 1), which we are developing in these days.


Example of a 3D web-gallery with 3DHOP (work in progress)

Being a Debian Jessie derivative distro, ArcheOS Hypatia will install 3DHOP in the folder /var/www/html/3dhop-3.0.
The best way to practice with this software is to modify the code of the different examples the package comes with (they are placed in /var/www/html/3dhop-3.0/examples).
For people that wants to help ArcheOS development and test this package, I upload it here. Soon we will prepare also 32-bit packages.

I hope this package will be useful! Merry Christmas and Happy new year!

PS

Marco Callieri (Visual Computing Lab - CNR-ISTI) notified me that Nexus and 3DHOP are under development right in these days, so there will be probably a new version in a short time. As soon as there will be changes, I will update the packages with the new functionalities.

Monday, 21 December 2015

Nexus 3 videotutorial

In order to help people in testing Nexus package for ArcheOS Hypatia (see the related post), I just recorded a fast videotutorial to show how this simple tool works.
Please note that the videotutorial refers to ArcheOS Hypatia's package, where Nexus binaries are stored in /usr/bin (so that they can be start from any path in the terminal). If you compiled your Nexus version somewhere in your home folder, than probably you will have to activate the binary from that directory.
Here below is the short video; I hope it will be useful.




Soon I will upload also the new 3DHOP package for ArcheOS Hypatia, with some instruction about it.
Have a nice day! 
BlogItalia - La directory italiana dei blog Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.