Showing posts with label pointcloud. Show all posts
Showing posts with label pointcloud. Show all posts

Monday, 27 July 2015

Documentation of a bas-relief on a cliff : the workflow

This summer, between May and June, we worked for a joint mission, led by the University of Innsbruck (Institut für Alte Geschichte und Altorientalistik) and the Cultural Heritage, Handcrafts and Tourism Organization of Iran. The project was held in Firuzabad, in the Pars Province of Iran. We will write more details about this work in the next post. By now I just want to use some material we collected to illustrate the work-flow in data acquiring during an archaeological documentation of a bas-relief on a cliff.
The video below shows the overall process.



You can see the initial preparation phase (1), during which we placed the Ground Control Point (GCP) to perform normal 2D vertical photo-mapping and to rectify and georeference the 3D point-cloud. Than (2) we collected pictures with three different flights of our DIY drone, in order to use them with different open source SfM/MVSR software (PPT, openMVG and MicMac), to reach the best possible result: a couple of flights with parallel camera, to have a good superimposition of the whole bas-relief, and a higher acquisition to cover the upper details. In the meantime (3) another operator (+Rupert Gietl) was collecting pictures from the ground, to register also the lower perspective. Later (4), I prepared the total station and collected the GCP, thanks to some fixed points we placed the day befor (0) with our GPS. Finally +Rupert Gietl  took the last (very close) details photos, using a ladder.
The entire process lasted more or less four hours, but we needed some more time the day before to place the fixed GCP down in the valley (in international Geographic Coordinates System). A good part of the work involved just the logistics or the approach to the site, and has been slowed by the transportation of the necessary equipment (ladder, total station and drone) through a couple of passages where it was necessary to climb some rocks.
It is interesting to note that it would not have been possible to accomplish this mission with a commercial drone, due to the embargo rules (which are currently under revision), while with a DIY hexacopter it has been simple to disassemble the components which were not allowed (like the FPV system ore the GPS controlled flight).
I hope this post was useful, have a nice day!

Tuesday, 16 June 2015

OpenMVG VS PPT

Hi all,
as I promised, we are back with new post in ATOR. We start today with an experiment we wanted to do since long: a comparison between two Structure from Motion - Muti View Stereoreconstruction (SfM - MVS) suite. The first is Python Photogrammetry Toolbox, developed by +Pierre Moulon some years ago and integrated in ArcheOS 4 (Caesar) with the new GUI (PPT-GUI) written in Python by +Alessandro Bezzi and me. The second one is the evolution of PPT: openMVG, which Pierre is developing since some years and that will be integrated in the next releases of ArcheOS.
Our small test regarded just four pictures taken with a Nikon D5000 on an old excavation. We want to point out the speed of the overall process in OpenMVG, which gave a result compatible with the one of PPT.
In the image below you can have an overview (in +MeshLab) of the two pointclouds generated bye the different software: openMVG processed a ply file with 391197 points, while PPT gave us a result with 425296 points.


Comparison of the two models generated by opnMVG and PPT

The main different stays in the processing time. In fact, while PPT needed 16 minutes, 11 seconds and 25 tenths, openMVG complete the model in just 3 minutes, 28 seconds and 20 tenths.
Here below I report the log file of openMVG, where you can see each step of the process:

STEP 1: Process Intrisic analysis - openMVG_main_CreateList took: 00:00:00:00.464
STEP 2: Process compute matches - openMVG_main_computeMatches took: 00:00:01:13.73764
STEP 3: Process Incremental Reconstruction -  openMVG_main_IncrementalSfM took: 00:00:00:47.47717
STEP 4: Process Export to CMVS-PMVS - openMVG_main_openMVG2PMVS took: 00:00:00:00.352
STEP 4: Process Export to CMVS-PMVS - openMVG_main_openMVG2PMVS took: 00:00:00:00.352
STEP 5: Process CMVS-PMVS took: 00:00:01:25.85958
--------------------
The whole detection and 3D reconsruction process took: 00:00:03:28.208258

We will go on in working and testing openMVG, hopfully posting soon news about this nice software.

Have a nice day!

Acknowledgment

Many thanks to +Pierre Moulon and +Cícero Moraes for the help!

Sunday, 15 February 2015

CloudCompare on Debian Wheezy by pinning from ArcheOS5 Theodoric


CloudCompare is a 3D point cloud processing software. It's deb is already packed for ArcheOS 5 Theodoric, under development on https://github.com/archeos/ArcheOS . To install on Debian Wheezy (7.8), just add theodoric's repo:
sudo nano /etc/apt/sources.list
then add this lines and save (ctrl+o, then ctrl+x):
#Archeos5 Theodoric deb http://repos.archeos.eu/apt theodoric main contrib non-free
To validate gpg's keys, write :
gpg --keyserver pgpkeys.mit.edu --recv-key 5AC5D028 gpg -a --export 5AC5D028 | sudo apt-key add -
and update source.list by:

sudo apt-get update
 
now you can able to install any software from theodoric's repo,  just pinning by "sudo apt-get install -t theodoric packet-name", let me show you an example that installs CloudCompare:
 sudo apt-get install -t theodoric cloudcompare

that's all.

Monday, 19 May 2014

MicMac and PPT: two FLOSS solutions for 3D data

Hi all,
last week I finished my lectures in the Master Open Techne about Free Software and 3D data (acquisition and processing). As last year I could spend many times to research new solutions and test some applications. Some months ago, thanks to the friend +Romain Janvier , I was introduce to the use of MicMac, a suite for three-dimensional documentation of reality developed by the Institut national de l’information géographique et forestière (IGN). As Python Photogrammetry Toolbox, MicMac can produce point cloud from set of photos. There are two different ways to acquire images:

- the ground geometry mode (useful for zenithal pictures as a drone data-set or for wall's prospect) = take pictures perpendicularly to the surface (ground or wall) with 60 % of overlapping (between images and lines of image)



- the image geometry mode (useful for any kind of object that has more than one surface) = take a "cross" of images starting from the central one and then up, down, left and right; then take other images moving to the second position (frontal to another surface) and again a "cross" of image; go on in this way for all the surfaces of the object that you need to reconstruct.



The data acquisition is a little bit more complicate than PPT, both in the way to shoot and in the camera settings (keep the same level of zoom, no auto-focus, no stabilization, no flash, ...), but the final point cloud is denser. PPT is more user-friendly (thank to the python scripts and the GUI) but slower in processing data (mostly in the Camera Pose Estimation step of Bundler).





One of the advantages of MicMac is the fast developing that is improving the software and simplifying its usage. I'm waiting for the GUI ;)
Unfortunately Bundler and CMVS/PMVS have not new release since years.

Monday, 25 February 2013

Cloud distance tool.

I was working on different SfM/IBM of a grave we dug in 2010. we have the documentation of four different levels (see picture below). It was a complex archaeological context, with two skeletons buried in different times (double burial), both partially destroyed by the construction of the Renaissance apse. Moreover the tomb was built on the side of a prehistoric house.



I tried to rectify the point clouds inside CloudCompare v. 2.4 (normally i use GRASS with the ply importer addon or MehLab) and I discover this fantastic tool: compute cloud/cloud distance. It can calculate the distance between two different overlapping mesh, similarly to the GRASS command "r.mapcalc". As you can see in the pictures below, the distance analysis between the first and the last documentation can represent the quantity of removed ground. It could be really useful for analysis of damages in buildings.

first point cloud

fourth point cloud

cloud/cloud distance

cloud/cloud distance over the fourth point cloud


BlogItalia - La directory italiana dei blog Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.