International and national journals
A. Cumani, A. Guiducci
Robot localisation error reduction by stereo vision
WSEAS Transactions on Circuits and Systems, Vol. 4, no. 10, pp. 1239-1245, October 2005.
In this work we deal with the problem of the long-term accumulation of pose errors in a grid-based SLAM system using a stereo vision sensor. We propose to periodically estimate the heading change by registering of tracked visual features. This approach allows to recover most of the heading error with respect to dead reckoning, while the 2D positional error is efficiently recovered by standard 2D map correlation. The results of some actual SLAM experiments are also shown, confirming the validity of the approach.
A. Chimienti, P. Dalmasso, R. Nerino, G. Pettiti, M. Spertino
A multiscale volumetric approach to surface reconstruction
WSEAS Transactions on Circuits and Systems, Vol. 4, no. 10, pp. 1268-1275, October 2005.
The inference of a three-dimensional (3D) surface from its sparse data samples is one of the most challenging problems in computer vision and in computer graphics. This paper describes a surface reconstruction method based on a volumetric approach where the reconstruction is obtained by building a sequence of surfaces approximating the data at increasing level of details. The method is based on a hierarchical partitioning of the volume data set. The working volume is split and classified at different scales of spatial resolution into surface, internal and external voxels, and this hierarchy is represented in a multiscale framework by an octree structure.
A volumetric description of the surface is build from the octree structure by representing the octree voxels as volumetric Radial Basis Functions (RBFs) of compact support. A hierarchy of surface approximations at different levels of details is then built from the octree scales. At each scale, information related to the reconstruction error drives the reconstruction process at the following finer scale. Differently from other volumetric methods, this method does not require information on surface normals. This aspect is important in the case of noisy data sets, such as those coming from image processing, because in this case normals are often estimated unreliably from the data. The method is simple, fast and suitable for a progressive 3D data/model representation, archiving, transmission. Preliminary results on synthetic and real data are presented and discussed.
Conference Proceedings
P. Grattoni, R. Nerino, G. Pettiti
Works of art monitoring by an active vision system
Proceedings of the 1st International Conference on "Science and Technology in Archeology and Conservation", pp. 383-391, Amman - Zarqa (Jordan), August 2002.
In the divulgation and preservation of cultural heritage important tasks are the acquisition and restoration processes. In-field monitoring of the degrade of works of art is the only feasible way to verify their degrade status, as in the case of monuments and frescoes where the object cannot be moved. In these cases the monitoring process is expensive, time consuming and it requires skilled operators. In this paper we present a new approach to this task which is based on an active stereo vision system. The system is able to acquire both the photometric/colorimetric (texture) and geometric (3D surfaces) information content of work of arts e.g. frescoes and monuments, with high dimensional and photometric accuracy. The system mimics the stereo fixation process of human eyes to localize objects in 3D thanks to its 3 degrees of freedom (two pan an one tilt) and is equipped with one wide angle camera for scene analysis and two motorized tele-camera for high resolution spatial localization and acquisition of 3D surface patches. The system features are described and typical applications to works of art monitoring outlined.
A. Cumani, A. Guiducci
Improving mobile robot localisation and map building by stereo vision
Proceedings of the 15th International Symposium on Measurement and Control in Robotics (ISMCR 2005), Brussels (Belgium), November 2005 (on cd rom).
In this paper, a method is proposed for reducing the long-term pose error in a grid-based SLAM system, using a panning stereo head. It is shown that, by periodically estimating the heading change by vision, most of the heading error with respect to dead reckoning can be recovered. The positional error can be efficiently recovered by standard 2D map correlation. Experimental results are shown, confirming the effectiveness of the approach.
A. Cumani, A. Guiducci
Mobile robot localisation with stereo vision
Proceedings of the 5th WSEAS International Conference on Signal Processing, Computational Geometry and Artificial Vision (ISCGAV'05), Malta, September 2005 (on cd rom).
This work considers the problem of reducing the accumulated pose error in a grid-based SLAM system using a stereo vision sensor. It is shown that by periodically estimating the heading change by vision it is possible to recover most of the heading error with respect to dead reckoning, while 2D positional error can be efficiently recovered by map correlation. Experimental results confirm the validity of the approach.
A. Chimienti, P. Dalmasso, R. Nerino, G. Pettiti, M. Spertino
Surface reconstruction from sparse data by a multiscale volumetric approach
Proceedings of 5th WSEAS ISCGAV' 05, Malta, September 2005 (on cd rom).
This paper describes a method for surface reconstruction from sparse three dimensional (3D) data that performs the reconstruction by building a sequence of surfaces approximating the data at increasing level of details (LOD). The method is simple, fast and suitable for a progressive 3D data/model representation, archiving, transmission. The surface reconstruction is obtained by a volumetric method that differs from other volumetric methods because it does not require implicitly or explicitly information on surface normals. This aspect is important in the case of noisy data sets, such as those coming from image based methods, because normals are often estimated unreliably from 3D data. The method is based on a hierarchical partitioning of the volume data set. The working volume is split and classified at different scales of spatial resolution into surface, internal and external voxels and this hierarchy is described by an octree structure in a multiscale framework. The octree structure is used to build a multiresolution description of the surface by means of compact support Radial Basis Functions (RBFs). A hierarchy of surface approximations at different LOD is built by representing the voxels at the same octree level as RBFs having the same spatial support. At each scale, information related to the reconstruction error drives the reconstruction process at the following finer scale. Preliminary results on real data are presented and discussed.
A. Balsamo, A. Chimienti, S. Desogus, P. Grattoni, A. Meda, R. Nerino, G. Pettiti, M. L. Rastello, M. Spertino
A portable stereovision system for Cultural Heritage monitoring
Annals of the CIRP, Vol. 54/1/2005, pp. 499-502.
In a nation-wide project named SIINDA, a novel instrument was developed and tested, intended for monitoring and diagnosing monuments. The instrument -named AVS- is portable to operate on field, and is capable of measuring 3D and colorimetric coordinates simultaneously. The paper describes the AVS, with a focus on its geometric measurement capability. Particular attention is given to the software compensation of the geometrical errors: the model is given, and the experimental plan to derive the model parameters is described, including the artefact designed and made for this purpose. Experimental results of laboratory and of on field tests are reported.
A. Balsamo, A. Chimienti, S. Desogus, P. Grattoni, G. Pettiti, A. Meda, R. Nerino, M. L. Rastello, M. Spertino
Il sistema di stereovisione attiva (SVA): uno strumento di misura integrato per i beni culturali
Atti del IV Congresso Metrologia e Qualità, pp. 265-268, Torino, febbraio 2005.
L'Italia è famosa nel mondo per la sua ricchezza di beni culturali, arte e monumenti. Se da un lato ciò costituisce certamente una ricchezza ed un'opportunità, dall'altra richiede un grande sforzo di conservazione, restauro e valorizzazione. Per questa ragione, v'è una crescente attenzione di ricerca su quest'argomento, sia pubblica sia privata. In questo quadro, s'è da poco concluso il progetto denominato SIINDA (Sistemi Innovativi d'INdagine e Diagnosi Assistita, vedi tabella 1) che ha avuto per oggetto lo sviluppo di un sistema integrato di conoscenza per la documentazione, il monitoraggio e la diagnosi di monumenti; il Teatro Romano d'Aosta è stato preso come caso di studio. All'interno di questo progetto, si è evidenziata la necessità di sviluppare uno strumento di misura portabile, in grado di misurare contemporaneamente le coordinate sia geometriche sia colorimetriche di opere d'arte, anche di monumenti posti all'aperto. Tale necessità nasce dal fatto che per l'analisi del degrado sono importanti sia la misura colorimetrica (utile ad esempio a discriminare depositi e ossidazioni) sia quella geometrica (ad esempio per la rilevazione d'erosioni in conci). Inoltre, è necessario sovrapporre le misurazioni colorimetriche effettuate in tempi successivi, ad esempio prima e dopo un intervento di restauro per valutarne l'efficacia, e ciò richiede la conoscenza del posizionamento nello spazio delle immagini, anche a distanza di mesi o anni. Si è quindi sviluppato e collaudato uno strumento idoneo allo scopo, che, per il suo principio di funzionamento, prende il nome di SVA, Sistema di Visione Attiva; esso è descritto nel seguito di questa memoria.