Path planning for image based visual servoing

This thesis deals with Visual Servoing and its strictly connected disciplines like projective geometry, image processing, robotics and non-linear control. More specifically the work addresses the problem to control a robotic manipulator through one of the largely used Visual Servoing techniques: the Image Based Visual Servoing (IBVS). In Image Based Visual Servoing the robot is driven by on-line performing a feedback control loop that is closed directly in the 2D space of the camera sensor. The work considers the case of a monocular system with the only camera mounted on the robot end effector (eye in hand configuration). Through IBVS the system can be positioned with respect to a 3D fixed target by minimizing the differences between its initial view and its goal view, corresponding respectively to the initial and the goal system configurations: the robot Cartesian Motion is thus generated only by means of visual informations. However, the execution of a positioning control task by IBVS is not straightforward because singularity problems may occur and local minima may be reached where the reached image is very close to the target one but the 3D positioning task is far from being fulfilled: this happens in particular for large camera displacements, when the the initial and the goal target views are noticeably different. To overcame singularity and local minima drawbacks, maintaining the good properties of IBVS robustness with respect to modeling and camera calibration errors, an opportune image path planning can be exploited. This work deals with the problem of generating opportune image plane trajectories for tracked points of the servoing control scheme (a trajectory is made of a path plus a time law). The generated image plane paths must be feasible i.e. they must be compliant with rigid body motion of the camera with respect to the object so as to avoid image jacobian singularities and local minima problems. In addition, the image planned trajectories must generate camera velocity screws which are smooth and within the allowed bounds of the robot. We will show that a scaled 3D motion planning algorithm can be devised in order to generate feasible image plane trajectories. Since the paths in the image are off-line generated it is also possible to tune the planning parameters so as to maintain the target inside the camera field of view even if, in some unfortunate cases, the feature target points would leave the camera images due to 3D robot motions. To test the validity of the proposed approach some both experiments and simulations results have been reported taking also into account the influence of noise in the path planning strategy. The experiments have been realized with a 6DOF anthropomorphic manipulator with a fire-wire camera installed on its end effector: the results demonstrate the good performances and the feasibility of the proposed approach.

Kryoelektronenmikroskopie von Proteinen: 3D-Struktur (12 A) des Hämocyanins der Schnecke Haliotis

Zusammenfassung:Die Quartärstruktur des respiratorischen Proteins Hämocyanin (Isoform HtH1) aus der marinen Schnecke Haliotis tuberculata wurde vermittels Kryoelektronen-mikroskopie und 3D-Rekonstruktion untersucht. Das Molekül ist zylinderförmig, hat einen Durchmesser von ca. 35 nm und besteht aus einer Zylinderwand und einem internen Kragenkomplex. Dieser wiederum besteht aus einem Collar und einem Arc.Die kryoelektronenmikroskopischen Aufnahmen von in glasartigem Eis fixierten HtH1-Molekülen brachte eine enorme Verbesserung der Anzahl der zur Verfügung stehenden Ansichtswinkel gegenüber den negativkontrastierten Molekülen, die auf Karbonfilm präpariert waren.Die 3D-Rekonstruktion des HtH1 mittels Aufnahmen bei drei verschiedenen Defo-kuswerten verbesserte die Auflösung noch einmal deutlich gegenüber den Rekon-struktionen, die aus Aufnahmen bei einem festen Defokuswert gemacht wurden, und zwar auf 12 Å. Das Molekül besitzt eine D5-Symmetrie.Aus dieser bisher genausten Rekonstruktion eines Molluskenhämocyanins aus EM-Bildern ließen sich folgende neue Strukturdetails ableiten:· Ein Untereinheitendimer konnte als Repeating Unit im Dekamer des HtH1 beschrieben werden.· Das Untereinheitendimer konnte aus der 3D-Dichtekarte isoliert werden. Es be-steht eindeutig aus 16 Massen, die funktionellen Domänen entsprechen. Zwei dieser Massen bilden den Collar, zwei den Arc und 12 das Wandsegment.· Die gegenläufige Anordnung der beiden Untereinheiten innerhalb dieses Unte-reinheitendimers konnten bestätigt und auf zwei Möglichkeiten eingeschränkt werden.· Die Zahl der alternativen Anordnungen der 16 funktionellen Domänen (HtH1-a bis HtH1-h) im Untereinheitendimer konnten von 80 auf 2 eingeengt werden.· Es konnte über molekulares Modellieren mithilfe einer publizierten Kristallstruk-tur eine 3D-Struktur fastatomarer Auflösung der funktionellen Domäne HtH1-g berechnet werden.· Die funktionelle Domäne HtH1-g konnte als Domänenpaar plausibel in die 3D?Dichtekarte des Untereinheitendimers eingepasst werden, und zwar in die beiden Massen des Arc.Aus der elektronenmikroskopisch gewonnenen Dichtekarte wurde mit Hilfe des

Valorizzazione di cartografia storica attraverso moderne tecniche geomatiche: recupero metrico, elaborazione e consultazione in ambiente digitale

“Cartographic heritage” is different from “cartographic history”. The second term refers to the study of the development of surveying and drawing techniques related to maps, through time, i.e. through different types of cultural environment which were background for the creation of maps. The first term concerns the whole amount of ancient maps, together with these different types of cultural environment, which the history has brought us and which we perceive as cultural values to be preserved and made available to many users (public, institutions, experts). Unfortunately, ancient maps often suffer preservation problems of their analog support, mostly due to aging. Today, metric recovery in digital form and digital processing of historical cartography allow preserving map heritage. Moreover, modern geomatic techniques give us new chances of using historical information, which would be unachievable on analog supports. In this PhD thesis, the whole digital processing of recovery and elaboration of ancient cartography is reported, with a special emphasis on the use of digital tools in preservation and elaboration of cartographic heritage. It is possible to divide the workflow into three main steps, that reflect the chapter structure of the thesis itself: • map acquisition: conversion of the ancient map support from analog to digital, by means of high resolution scanning or 3D surveying (digital photogrammetry or laser scanning techniques); this process must be performed carefully, with special instruments, in order to reduce deformation as much as possible; • map georeferencing: reproducing in the digital image the native metric content of the map, or even improving it by selecting a large number of still existing ground control points; this way it is possible to understand the projection features of the historical map, as well as to evaluate and represent the degree of deformation induced by the old type of cartographic transformation (that can be unknown to us), by surveying errors or by support deformation, usually all errors of too high value with respect to our standards; • data elaboration and management in a digital environment, by means of modern software tools: vectorization, giving the map a new and more attractive graphic view (for instance, by creating a 3D model), superimposing it on current base maps, comparing it to other maps, and finally inserting it in GIS or WebGIS environment as a specific layer. The study is supported by some case histories, each of them interesting from the point of view of one digital cartographic elaboration step at least. The ancient maps taken into account are the following ones: • three maps of the Po river delta, made at the end of the XVI century by a famous land-surveyor, Ottavio Fabri (he is single author in the first map, co-author with Gerolamo Pontara in the second map, co-author with Bonajuto Lorini and others in the third map), who wrote a methodological textbook where he explains a new topographical instrument, the squadra mobile (mobile square) invented and used by himself; today all maps are preserved in the State Archive of Venice; • the Ichnoscenografia of Bologna by Filippo de’ Gnudi, made in the 1702 and today preserved in the Archiginnasio Library of Bologna; it is a scenographic view of the city, captured in a bird’s eye flight, but also with an icnographic value, as the author himself declares; • the map of Bologna by the periti Gregorio Monari and Antonio Laghi, the first map of the city derived from a systematic survey, even though it was made only ten years later (1711–1712) than the map by de’ Gnudi; in this map the scenographic view was abandoned, in favor of a more correct representation by means of orthogonal projection; today the map is preserved in the State Archive of Bologna; • the Gregorian Cadastre of Bologna, made in 1831 and updated until 1927, now preserved in the State Archive of Bologna; it is composed by 140 maps and 12 brogliardi (register volumes). In particular, the three maps of the Po river delta and the Cadastre were studied with respect to their acquisition procedure. Moreover, the first maps were analyzed from the georeferencing point of view, and the Cadastre was analyzed with respect to a possible GIS insertion. Finally, the Ichnoscenografia was used to illustrate a possible application of digital elaboration, such as 3D modeling. Last but not least, we must not forget that the study of an ancient map should start, whenever possible, from the consultation of the precious original analogical document; analysis by means of current digital techniques allow us new research opportunities in a rich and modern multidisciplinary context.

Vergleichende computergestützte funktionsmorphologische Analyse an Molaren cercopithecoider Primaten

Die Analyse funktioneller Zusammenhänge zwischen Ernährung und Zahnmorphologie ist ein wichtiger Aspekt primatologischer und paläontologischer Forschung. Als überdauernder Teil des Verdauungssystems geben Zähne die bestmöglichen Hinweise auf die Ernährungsstrategien (ausgestorbener) Arten und eine Fülle weiterer Informationen. Aufgrund dessen ist es für die wissenschaftliche Arbeit von größter Bedeutung, die Zähne so detailliert und exakt wie möglich in ihrer gesamten Struktur zu erfassen. Bisher wurden zumeist zweidimensionale Parameter verwendet, um die komplexe Kronenmorphologie von Primatenmolaren vergleichend zu untersuchen. Die vorliegende Arbeit hatte das Ziel, Zähne verschiedener Arten von Altweltaffen mittels computerbasierter Methoden dreidimensional zu erfassen und neue Parameter zu definieren, mit denen die Form dieser Zähne objektiv erfasst und funktionell interpretiert werden kann. Mit einem Oberflächen-Scanner wurden die Gebisse einer Stichprobe von insgesamt 48 Primaten von fünf verschiedenen Arten eingescannt und mit Bildverarbeitungsmethoden so bearbeitet, dass dreidimensionale digitale Modelle einzelner Backenzähne zur Analyse vorlagen. Es wurden dabei sowohl Arten ausgewählt, die eine für ihre Gattung typische Ernährungsweise besitzen - also Frugivorie bei den Cercopithecinen und Folivorie bei den Colobinen - als auch solche, die eine davon abweichende Alimentation bevorzugen. Alle Altweltaffen haben sehr ähnliche Molaren. Colobinen haben jedoch höhere und spitzere Zahnhöcker, dünneren Zahnschmelz und scheinen ihre Zähne weniger stark abzukauen als die Meerkatzen. Diese Beobachtungen konnten mit Hilfe der neuen Parameter quantifiziert werden. Aus der 3D-Oberfläche und der Grundfläche der Zähne wurde ein Index gebildet, der die Stärke des Oberflächenreliefs angibt. Dieser Index hat bei Colobinen deutlich höhere Werte als bei Cercopithecinen, auch bei Zähnen, die schon stark abgekaut sind. Die Steilheit der Höcker und ihre Ausrichtung wurden außerdem gemessen. Auch diese Winkelmessungen bestätigten das Bild. Je höher der Blätteranteil an der Ernährung ist, desto höher sind die Indexwerte und umso steiler sind die Höcker. Besonders wichtig war es, dies auch für abgekaute Zähne zu bestätigen, die bisher nicht in funktionelle Analysen miteinbezogen wurden. Die Ausrichtung der Höckerseiten gibt Hinweise auf die Kaubewegung, die zum effizienten Zerkleinern der Nahrung notwendig ist. Die Ausrichtung der Höcker der Colobinen deutet darauf hin, dass diese Primaten flache, gleitende Kaubewegungen machen, bei denen die hohen Höcker aneinander vorbei scheren. Dies ist sinnvoll zum Zerschneiden von faserreicher Nahrung wie Blättern. Cercopithecinen scheinen ihre Backenzähne eher wie Mörser und Stößel zu verwenden, um Früchte und Samen zu zerquetschen und zu zermahlen. Je nachdem, was neben der hauptsächlichen Nahrung noch gekaut wird, unterscheiden sich die Arten graduell. Anders als bisher vermutet wurde, konnte gezeigt werden, dass Colobinen trotz des dünnen Zahnschmelzes ihre Zähne weniger stark abkauen und weniger Dentin freigelegt wird. Dies gibt eindeutige Hinweise auf die Unterschiede in der mechanischen Belastung, die während des Kauvorgangs auf die Zähne wirkt, und lässt sich gut mit der Ernährung der Arten in Zusammenhang bringen. Anhand dieser modellhaften Beobachtungen können in Zukunft ausgestorbene Arten hinsichtlich ihrer Ernährungsweise mit 3D-Techniken untersucht werden.

Laser scanning microscopy combined with image restoration to analyse a 3D model of the human epithelial airway barrier

A laser scanning microscope collects information from a thin, focal plane and ignores out of focus information. During the past few years it has become the standard imaging method to characterise cellular morphology and structures in static as well as in living samples. Laser scanning microscopy combined with digital image restoration is an excellent tool for analysing the cellular cytoarchitecture, expression of specific proteins and interactions of various cell types, thus defining valid criteria for the optimisation of cell culture models. We have used this tool to establish and evaluate a three dimensional model of the human epithelial airway wall.

Wow! Linear systems and signal processing is fun!

We describe a recent offering of a linear systems and signal processing course for third-year electrical and computer engineering students. This course is a pre-requisite for our first digital signal processing course. Students have traditionally viewed linear systems courses as mathematical and extremely difficult. Without compromising the rigor of the required concepts, we strived to make the course fun, with application-based hands-on laboratory projects. These projects can be modified easily to meet specific instructors' preferences. © 2011 IEEE.(17 refs)

The discrete Pascal transform and its applications

We introduce a new discrete polynomial transform constructed from the rows of Pascal’s triangle. The forward and inverse transforms are computed the same way in both the oneand two-dimensional cases, and the transform matrix can be factored into binary matrices for efficient hardware implementation. We conclude by discussing applications of the transform in

Distance measures for image segmentation evaluation

The task considered in this paper is performance evaluation of region segmentation algorithms in the ground-truth-based paradigm. Given a machine segmentation and a ground-truth segmentation, performance measures are needed. We propose to consider the image segmentation problem as one of data clustering and, as a consequence, to use measures for comparing clusterings developed in statistics and machine learning. By doing so, we obtain a variety of performance measures which have not been used before in image processing. In particular, some of these measures have the highly desired property of being a metric. Experimental results are reported on both synthetic and real data to validate the measures and compare them with others.

Non-invasive method to predict viscosity and size using total internal reflection fluorescence microscopy (TIRFM) system

This study develops an automated analysis tool by combining total internal reflection fluorescence microscopy (TIRFM), an evanescent wave microscopic imaging technique to capture time-sequential images and the corresponding image processing Matlab code to identify movements of single individual particles. The developed code will enable us to examine two dimensional hindered tangential Brownian motion of nanoparticles with a sub-pixel resolution (nanoscale). The measured mean square displacements of nanoparticles are compared with theoretical predictions to estimate particle diameters and fluid viscosity using a nonlinear regression technique. These estimated values will be confirmed by the diameters and viscosities given by manufacturers to validate this analysis tool. Nano-particles used in these experiments are yellow-green polystyrene fluorescent nanospheres (200 nm, 500 nm and 1000 nm in diameter (nominal); 505 nm excitation and 515 nm emission wavelengths). Solutions used in this experiment are de-ionized (DI) water, 10% d-glucose and 10% glycerol. Mean square displacements obtained near the surface shows significant deviation from theoretical predictions which are attributed to DLVO forces in the region but it conforms to theoretical predictions after ~125 nm onwards. The proposed automation analysis tool will be powerfully employed in the bio-application fields needed for examination of single protein (DNA and/or vesicle) tracking, drug delivery, and cyto-toxicity unlike the traditional measurement techniques that require fixing the cells. Furthermore, this tool can be also usefully applied for the microfluidic areas of non-invasive thermometry, particle tracking velocimetry (PTV), and non-invasive viscometry.

DIMENSION REDUCTION FOR POWER SYSTEM MODELING USING PCA METHODS CONSIDERING INCOMPLETE DATA READINGS

Principal Component Analysis (PCA) is a popular method for dimension reduction that can be used in many fields including data compression, image processing, exploratory data analysis, etc. However, traditional PCA method has several drawbacks, since the traditional PCA method is not efficient for dealing with high dimensional data and cannot be effectively applied to compute accurate enough principal components when handling relatively large portion of missing data. In this report, we propose to use EM-PCA method for dimension reduction of power system measurement with missing data, and provide a comparative study of traditional PCA and EM-PCA methods. Our extensive experimental results show that EM-PCA method is more effective and more accurate for dimension reduction of power system measurement data than traditional PCA method when dealing with large portion of missing data set.

Dual-Domain Image Denoising

Image denoising methods have been implemented in both spatial and transform domains. Each domain has its advantages and shortcomings, which can be complemented by each other. State-of-the-art methods like block-matching 3D filtering (BM3D) therefore combine both domains. However, implementation of such methods is not trivial. We offer a hybrid method that is surprisingly easy to implement and yet rivals BM3D in quality.

Quantitative 3D strain analysis in analogue experiments: Integration of X-ray computed tomography and digital volume correlation techniques

The combination of scaled analogue experiments, material mechanics, X-ray computed tomography (XRCT) and Digital Volume Correlation techniques (DVC) is a powerful new tool not only to examine the 3 dimensional structure and kinematic evolution of complex deformation structures in scaled analogue experiments, but also to fully quantify their spatial strain distribution and complete strain history. Digital image correlation (DIC) is an important advance in quantitative physical modelling and helps to understand non-linear deformation processes. Optical non-intrusive (DIC) techniques enable the quantification of localised and distributed deformation in analogue experiments based either on images taken through transparent sidewalls (2D DIC) or on surface views (3D DIC). X-ray computed tomography (XRCT) analysis permits the non-destructive visualisation of the internal structure and kinematic evolution of scaled analogue experiments simulating tectonic evolution of complex geological structures. The combination of XRCT sectional image data of analogue experiments with 2D DIC only allows quantification of 2D displacement and strain components in section direction. This completely omits the potential of CT experiments for full 3D strain analysis of complex, non-cylindrical deformation structures. In this study, we apply digital volume correlation (DVC) techniques on XRCT scan data of “solid” analogue experiments to fully quantify the internal displacement and strain in 3 dimensions over time. Our first results indicate that the application of DVC techniques on XRCT volume data can successfully be used to quantify the 3D spatial and temporal strain patterns inside analogue experiments. We demonstrate the potential of combining DVC techniques and XRCT volume imaging for 3D strain analysis of a contractional experiment simulating the development of a non-cylindrical pop-up structure. Furthermore, we discuss various options for optimisation of granular materials, pattern generation, and data acquisition for increased resolution and accuracy of the strain results. Three-dimensional strain analysis of analogue models is of particular interest for geological and seismic interpretations of complex, non-cylindrical geological structures. The volume strain data enable the analysis of the large-scale and small-scale strain history of geological structures.

Progressive Image Denoising

Image denoising continues to be an active research topic. Although state-of-the-art denoising methods are numerically impressive and approch theoretical limits, they suffer from visible artifacts.While they produce acceptable results for natural images, human eyes are less forgiving when viewing synthetic images. At the same time, current methods are becoming more complex, making analysis, and implementation difficult. We propose image denoising as a simple physical process, which progressively reduces noise by deterministic annealing. The results of our implementation are numerically and visually excellent. We further demonstrate that our method is particularly suited for synthetic images. Finally, we offer a new perspective on image denoising using robust estimators.

Image analysis of immunohistochemistry is superior to visual scoring as shown for patient outcome of esophageal adenocarcinoma.

Quantification of protein expression based on immunohistochemistry (IHC) is an important step in clinical diagnoses and translational tissue-based research. Manual scoring systems are used in order to evaluate protein expression based on staining intensities and distribution patterns. However, visual scoring remains an inherently subjective approach. The aim of our study was to explore whether digital image analysis proves to be an alternative or even superior tool to quantify expression of membrane-bound proteins. We analyzed five membrane-binding biomarkers (HER2, EGFR, pEGFR, β-catenin, and E-cadherin) and performed IHC on tumor tissue microarrays from 153 esophageal adenocarcinomas patients from a single center study. The tissue cores were scored visually applying an established routine scoring system as well as by using digital image analysis obtaining a continuous spectrum of average staining intensity. Subsequently, we compared both assessments by survival analysis as an end point. There were no significant correlations with patient survival using visual scoring of β-catenin, E-cadherin, pEGFR, or HER2. In contrast, the results for digital image analysis approach indicated that there were significant associations with disease-free survival for β-catenin, E-cadherin, pEGFR, and HER2 (P = 0.0125, P = 0.0014, P = 0.0299, and P = 0.0096, respectively). For EGFR, there was a greater association with patient survival when digital image analysis was used compared to when visual scoring was (visual: P = 0.0045, image analysis: P < 0.0001). The results of this study indicated that digital image analysis was superior to visual scoring. Digital image analysis is more sensitive and, therefore, better able to detect biological differences within the tissues with greater accuracy. This increased sensitivity improves the quality of quantification.

Is complex signal processing for bone conduction hearing aids useful?

OBJECTIVES To establish whether complex signal processing is beneficial for users of bone anchored hearing aids. METHODS Review and analysis of two studies from our own group, each comparing a speech processor with basic digital signal processing (either Baha Divino or Baha Intenso) and a processor with complex digital signal processing (either Baha BP100 or Baha BP110 power). The main differences between basic and complex signal processing are the number of audiologist accessible frequency channels and the availability and complexity of the directional multi-microphone noise reduction and loudness compression systems. RESULTS Both studies show a small, statistically non-significant improvement of speech understanding in quiet with the complex digital signal processing. The average improvement for speech in noise is +0.9 dB, if speech and noise are emitted both from the front of the listener. If noise is emitted from the rear and speech from the front of the listener, the advantage of the devices with complex digital signal processing as opposed to those with basic signal processing increases, on average, to +3.2 dB (range +2.3 … +5.1 dB, p ≤ 0.0032). DISCUSSION Complex digital signal processing does indeed improve speech understanding, especially in noise coming from the rear. This finding has been supported by another study, which has been published recently by a different research group. CONCLUSIONS When compared to basic digital signal processing, complex digital signal processing can increase speech understanding of users of bone anchored hearing aids. The benefit is most significant for speech understanding in noise.