Geoscience Reference
In-Depth Information
at 21 %. So, not only 3D metadata is not common with regard to the attributed
concepts, definitions, and terms, it also depends on sparse individual and organiza-
tional initiatives (with most of the information found in written abstracts and sum-
mers). The latter was also a challenge for carrying out the inventory as it required
investigating several number of metadata documents generated by distinct ration-
alities and writing skills.
In response, our goal was to find the generic requirements for documenting
3D geospatial data and to model them at higher level of abstraction than being
domain and application specific. The requirements were identified by studying
the concepts by which the academia and actual practitioners outline their 3D
resources. These requirements were cited to propose our 3D metadata model
which integrates the contextual and structural specifications of 3D geospatial
data with as many explicit classes as possible. For first assessments, the pro-
posed 3D metadata was compared with Canada's CGDI Discovery Portal meta-
data which partially conforms to North American Profile (NAP) of ISO 19115
and was the most explicit inventoried metadata (Table
1
). At conceptual level,
the proposed 3D metadata brings up six new concepts for documenting 3D geo-
spatial datasets. This corresponds to a 42 % jump in the explicitness of meta-
data for being used with 3D data. To compare the suitability of both metadata
with regards to the practical cases which seek 3D geospatial data, three simu-
lated scenarios were established about discovering specific types of 3D datasets.
These scenarios were parameterized based the respective publications and indus-
trial preferences. Our proposal shows promising results as it is noticed beneficial
by being extensive and explicit on the definition and conceptualization of the 3rd
dimension, 3D geometric modeling methods and structures, vertical referencing,
and various aspects of LoD.
Currently, all of the proposed metadata classes which are directly aggregated
to the
MD_TRGT
core class of the proposed 3D metadata (Fig.
2
) are manda-
tory. It should be required in future works to constrain the proposed metadata
classes according to users' preferences. The proposed 3D metadata is quite tech-
nical on some of the proposed domain values. Although this helps with seman-
tic coherence, further work is certainly required to make the proposed domain
values more accessible (i.e. covering the technical expressions while keeping
the domain values simple for public). In such sense, working on an ontological
expression of 3D knowledge would be recommended. The mentioned amend-
ments are important to avoid redundancy in the documented 3D information as
well as the probable confusions among non-expert users when encountering the
proposed 3D metadata.
Now, the proposed 3D metadata needs to be validated at larger scale. In such
direction, an interesting future work is to analyse the proposed 3D metadata
against the actual queries made about 3D geospatial models in relevant geo-portals
such as Canada's CGDI Discovery Portal, Princeton University 3D Model Search
Engine, and Turbo Squid Portal. The future works also includes receiving the
recognition of the beneficiary stakeholders by implementing a prototype system
which enables the proposed 3D metadata for public use.
