1,721,103 research outputs found
Arothron: R Functions for Geometric Morphometrics Analyses
No full text<p>Tools for Geometric Morphometrics</p>
Arothron: R Functions for Geometric Morphometrics Analyses
No full text<p>Tools for Geometric Morphometrics</p>
Arothron: R Functions for Geometric Morphometrics Analyses
No full textTools for Geometric Morphometric
Advances in virtual morphometrics: a new approach to generate 2D and 3D surface outlines on virtual specimens
No full textPalaeoanthropology and Bioarchaeology make extensive use of outline-based morphometrics and this approach can be used to ex-tract information from several sources. 2D outlines are usually generated from teeth or lithic tools, as well as from skulls or longbones profiles, and this method is widespread in the study of human evolution [1, 2]. In many cases, the generation of outlines is avaluable alternative to traditional linear measurements or 2D landmark-based geometric morphometrics. 2D outlines are conven-tionally acquired through photography, a procedure prone to parallax errors . In recent years, the use of virtual collections increasedremarkably due to the availability of novel imaging techniques, such as CT-scanning, laser scanning and photogrammetry. anksto these new virtual environments, current methodological approaches can be improved and new techniques can be developed. Wepresent here a method to generate 2D and 3D outlines of complex specimens from a 3D surface model. e 3D contours recreatethe patterns of maximal breadth of the object outline. e procedure has been developed in the R statistical environment and usesα-shape approach and Bézier curves to generate the outlines. α-shape formalises the concept of “shape” for spatial point sets in com-putational geometry : a α-shape is built by connecting all the pairs of points lying on a circle which is not touching or overlappingany other point of the set. Bézier curves are used to approximate curves by generating a set of points through a polynomial fitting.Only three landmarks are needed to apply the method: these points define a reference plane to project all the vertices of the 3Dmodel onto. A α-shape is then obtained from the projected vertices and a Bézier approximation is used to generate evenly spacedlandmarks lying on the outline. e 3D outline is generated by bringing each point of the α-shape back on the surface model; aBézier curve is then calculated. is method allows high precision and reproducibility in outline generation and can be applied onboth skeletal and lithic material. 3D contours generated with this method consent to address new questions on functional trendsin skeletal morphology or in material culture, enhancing the morphometric approach on the study of human evolution
Old and new methods in geometric morphometrics applied to the study of human evolution: case studies
Full text linkThe main aim of this thesis is the application of Geometric Morphometric methods on different case-studies for the study of human evolution. When combined with techniques of acquisition of 3D models, Geometric Morphometrics allows us: - to investigate different experimental designs;
- to analyse the interaction of several aspects on morphological adaptations detected within taxon-specific studies (e.g., allometry, phylogenetic signal, functional factors);
- to study fragmentary fossils and incomplete specimens, through estimation of missing data and 3D virtual restoration (e.g., surface and curve slid semilandmark);
- to plan, code and test new algorithms and/or methodological approaches.
The first part deals with the use of smoothing filters applied to 3D model for Geometric Morphometric studies. These algorithms are used to remove the background noise deriving from digital acquisition (e.g., photogrammetry, laser scan and computerized tomography scan). The effects of the different smoothing filters have been assessed. In particular have been defined guidelines for a correct use of these algorithms, besides the developing of an automatized tool, in R environment, to find the best combination between algorithm type, settings and number of iterations.
The second part consists of a protocol developed, with the collaboration of the University of Freiburg, for the digital retrodeformation of fossil specimens showing evidence of shearing, bending and compressing alterations due to taphonomic processes. Traditional methods of retrodeformation only use a sparse set of bilateral landmarks; the number of points appears to affects the success of retrodeformation. On the contrary, this method uses, in addition to the landmark configurations, the curve and surface semilandmarks, which allow us to capture
morphological information more accurately. This protocol was applied here to the neanderthalian cranium of Saccopastore 1.
The third part reports the results of the first analysis on the specimen nicknamed "Pàus” (St.n.166623), recently discovered near Spinadesco in the Po Valley (Northern Italy). A set of 100 semilandmarks was built on the specimen and slid, using a set of 6 landmark, on a comparative sample including specimens dated to the Middle-to-Late Pleistocene. The results show how the morphology of “Pàus” is consistent with the variability observed in the Neanderthal lineage.
The fourth part concerns a Geometric Morphometric investigation performed on two human cranial fossil remains from Melka Kunture, dated to about 850 ka. The two cranial fragments consist in a partial left parietal (MK73/GOM II- 6769; formally Melka Kunture 1, or MK1) and a right portion of the frontal bone (MK76/GOM II - 576, or MK2). Specifically, evenly-spaced semilandmark sets were used acquired along the sagittal suture and the inferior temporal line on MK1 and MK2 respectively. The results of the analyses, in agreement with the chronology of the fossils, represent at present, evidence of one of the best candidates to be the most ancient example of H. heidelbergensis.
The fifth part treats with the external morphology of the cranial base in extant and living Hominoids in relation to ontogenetic, allometric, locomotor and phylogenetic factors. The sample selected consists of 3D landmark configurations acquired on specimens (male and female) including infant, sub-adult and adult individuals. The centroid size of the landmark set configuration was used as indicator of size while the pattern of dental eruption (at death) was used to define six age groups. The relation between morphology and locomotion was explored through the estimation of the position of the foramen magnum along the Frankfurt plane.
Finally, a phylogenetic tree was build using molecular and paleontological data, with the phylogenetic signal investigated through centroid size and shape
Arothron: Geometric Morphometrics Analyses
No full text<p>Tools for geometric morphometric analysis. The package includes tools of virtual anthropology to align two not articulated parts belonging to the same specimen and to build virtual cavities as endocast. In addition, we supply functions to import and export the coordinates of landmarks and 3D paths into 'landmarkAscii' and 'am' format files.</p>
Frontal bone virtual reconstruction and geometric morphometric analysis of the mid-Pleistocene hominin KNM-OG 45500 (Olorgesailie, Kenya)
Full text linkKNM-OG 45500 is a hominin fossil composed of parts of a frontal bone, left temporal bone, and cranial vault pieces. Since its discovery along the Olorgesailie Formation (Kenya) in 2003, it has been associated with the Homo erectus hypodigm. The specimen, derived from a geological context dated to ca. 900 Ka BP, has been described as a very small individual of probable female sex. However, despite its status as an important hominin specimen, it has not been used in a quantitative comparative framework because of its fragmentary condition. Here, we undertake a virtual reconstruction of the better-preserved fragment, the frontal bone. We additionally apply geometric morphometric analyses, using a geographically diverse fossil and modern human sample, in order to investigate the morphological affinities of KNM-OG 45500. Our results show that the frontal shape of KNM-OG 45500 exhibits similarities with Early Pleistocene fossils from Eurasia and Africa that are assigned to H. erectus sensu lato (s.l.). Its size, on the other hand, is notably smaller than most other Homo erectus fossils and modern humans and similar to the specimens from Dmanisi (Georgia) and to Homo naledi. Taken together, our analyses of the frontal bone suggest a taxonomic attribution of KNM-OG 45500 to H. erectus s.l. and extend even further the range of size variability associated with this taxon around 900 Ka BP
Digital alignment: an automatized protocol for virtual reconstruction of incomplete fossil specimens
No full textMuseal collections include a large amount of cranial specimens of living and fossil primates. This material served as the basis for
detailed morphological studies also relatively to their internal anatomy. Unfortunately, until recently, the study of the cranial cavity
(e.g. the endocranial cavity) was often possible only by removing mechanically extensive portions of the cranial vault and of other
skeletal portions, in fact compromising the physical integrity of the specimens. Many specimens have gone through fractures and
unintentional damage, which may also have caused the loss of original morphological information. Thanks to increasing advances in
both computer technology and 3D imaging software, it is possible to virtually acquire the morphology of a physical specimen [1,2].
Consequently, paleoanthropological studies often focus on anatomy, virtual reconstruction, and on the development of algorithms
to improve the digital acquisition [3]. In addition to CT scan other techniques have been introduced, such as laser scanner and photogrammetry.
Virtual procedures, on the basis of digital reconstruction, are frequently applied to restore human fossil specimens.
A digital operation on a 3D specimen is appropriate and/or necessary when the object is fragmented/damaged and/or deformed
by taphonomical pressures. Here we present a protocol, developed in R environment, able to align automatically two portions belonging
to the same 3D model. In this communication, we introduce the method and we report the results of the application of this
tool applied on a skull of Homo sapiens (target model). The target model was divided in two halves and each portion was shifted in
the xyz reference Cartesian system. The aim was to compare the efficacy of this protocol (full computer assisted alignment) with
a manual alignment. The digital alignment consists of the extrapolation of the rotation matrix to translate, rotate, and scale a fragment
of a target model using a reference model. In literature, almost all digital reconstructions of fossil specimens start from an
arbitrary model chosen as reference [4]. The first part of the tool is dedicated to the detection of the reference model using a landmark
configuration as guide. Once chosen, the reference model was symmetrized using a bilateral configuration landmark. The two
halves of the target model were aligned on the symmetrized version of the reference model, and optionally the alignment could be
corrected on the basis of an external ratios or angles (in this case the angle-glabella-inion). Finally, the alignment performed using
this protocol was compared with 10 manually-performed alignments carried out by 10 anonymous researchers expert in the field.
In this case the target model aligned automatically results to be the closer to the starting model; in fact the mesh distance values
is minor than alignments manually performed. In sum, the digital alignment of a fragmentary fossil specimen is the first stage of a
reconstruction procedure and the correct choice of the reference model is the crucial point of the digital reconstruction of a damaged
specimen. The application of a computer-assisted reconstruction implies that the efficiency of the reconstruction depends on
the comparative sample (3D models and landmark/semi-landmark sets) and not on the skills of the operator. This first release of
the tool will be fully open-access and available to the scientific community for application and methodological improvements. A
first application of this protocol is represented by the work done on the Altamura Neanderthal [5]
A Comparison of Semilandmarking Approaches in the Analysis of Size and Shape
Full text linkOften, few landmarks can be reliably identified in analyses of form variation and covariation. Thus, ‘semilandmarking’ algorithms have increasingly been applied to surfaces and curves. However, the locations of semilandmarks depend on the investigator’s choice of algorithm and their density. In consequence, to the extent that different semilandmarking approaches and densities result in different locations of semilandmarks, they can be expected to yield different results concerning patterns of variation and co-variation. The extent of such differences due to methodology is, as yet, unclear and often ignored. In this study, the performance of three landmark-driven semilandmarking approaches is assessed, using two different surface mesh datasets (ape crania and human heads) with different degrees of variation and complexity, by comparing the results of morphometric analyses. These approaches produce different semilandmark locations, which, in turn, lead to differences in statistical results, although the non-rigid semilandmarking approaches are consistent. Morphometric analyses using semilandmarks must be interpreted with due caution, recognising that error is inevitable and that results are approximations. Further work is needed to investigate the effects of using different landmark and semilandmark templates and to understand the limitations and advantages of different semilandmarking approaches
- …
