×

You are using an outdated browser that does not fully support the intranda viewer.
As a result, some pages may not be displayed correctly.

We recommend you use one of the following browsers:

Full text

Title
New perspectives to save cultural heritage
Author
Altan, M. Orhan

CIPA 2003 XIX"' International Symposium, 30 September - 04 October, 2003, Antalya, Turkey
and it is important that users are aware of the class of their
instrument in the required classification.
4.3 Data storage
No specific data format or standardised set of metadata exists
for TLS. It is not within the scope of this project to define a
suitable format and supply the necessary management
requirements. A brief overview of the issues involved
however would benefit discussion of the topic. The LAS
format (LAS, 2002) has recently been adopted by the ASPRS
as a standard for airborne laser scanning (ALS). Although
ALS and TLS share some common features and that the LAS
format could be easily adapted to store TLS data it is
suggested that a new format should be developed to ensure
full compatibility with all future software systems. Any such
format should be universally accepted by software
developers and users. The following issues should be
considered in any format definition:
■ To maintain simplicity a single file should
represent a single scan position (header files could
be used to group together individual scans).
■ Storing information in an arbitrary system would
seem good practice but transformation parameters
are also required so a single scan may be viewed
together with other scans in a “real world”
coordinate system.
■ Support must be provided to store intensity and/or
RGB values for each point.
■ Optimum compression of the data should be
possible. The data volumes provided by the
available systems vary, with the Z+F Imager 5003
providing up to 1GB from a single scan position.
A compressed binary format is therefore
preferable.
As no format is currently available it has been necessary to
adapt the specification to allow for a variety of formats with
the emphasis being placed on the ability to transfer data
between software systems. The transfer of data between
software systems would be best facilitated through the
application of a standard level of metadata, both for
individual scans and for a project as a whole. For raw scan
data the specification requires the following metadata:
■ Total number of points
■ Point spacing on the object
■ Filename for control data
■ Weather during survey
■ The file name of an image, located at the point of
collection, showing the data collected
This metadata should be contained in a digital file and in
hard copy in the post survey report. The post survey report
should also include the standard requirements found in the
current metric survey specification such as details of control
point coordinates and accuracies.
Although the generation of this information may at first seem
like additional work for the contractor it is envisaged that in
future all metadata would be generated semi-automatically.
Most systems for manipulating and processing scan data now
accepted control and image data in addition to scan data.
This could also be extended to contain notes and sketches.
Such data management systems (DMS) would make
metadata generation easier and in the long term these DMS
may become full GIS systems which advanced 3D spatial
data analysis tools. Such DMS may even become actual
deliverables themselves.
It is common to find users stating that decimation of data was
required before any viewing/processing could be performed.
It may be necessary to define appropriate decimation
procedures to allow for this if required. As the standard
computer specification increases the impact of this limitation
maybe reduced.
4.4 Common faults
Finally the specification aims to limit some of the common
faults of laser scanning surveys. These include:
■ Data voids - These are normally caused by
temporary/permanent occlusions of the
measurement beam by vehicle or pedestrian traffic
(Figure 3) or obstruction by the building itself.
Where it is not possible to prevent such occlusions
it will be necessary to provide sufficient overlap
between scans to fill such voids.
■ File name of the raw data
■ Date of capture
■ Scanning system used (with serial number)
■ Company name
■ Monument name
■ Monument number (if known)
■ Survey number (if known)
■ Scan number (unique scan number for this survey)