The spectral database SPECCHIO in support of Cal/Val activities

Loading.... (view fulltext now)





Full text


University of Zurich

Zurich Open Repository and Archive

Winterthurerstr. 190 CH-8057 Zurich

Year: 2010

The spectral database SPECCHIO in support of Cal/Val


Hueni, A; Kneubuehler, M

Hueni, A; Kneubuehler, M (2010). The spectral database SPECCHIO in support of Cal/Val activities. In: Hyperspectral Workshop 2010, Frascati, IT, 17 March 2010 - 19 March 2010, 1-5.

Postprint available at:

Posted at the Zurich Open Repository and Archive, University of Zurich.

Originally published at:

Hyperspectral Workshop 2010, Frascati, IT, 17 March 2010 - 19 March 2010, 1-5.

Hueni, A; Kneubuehler, M (2010). The spectral database SPECCHIO in support of Cal/Val activities. In: Hyperspectral Workshop 2010, Frascati, IT, 17 March 2010 - 19 March 2010, 1-5.

Postprint available at:

Posted at the Zurich Open Repository and Archive, University of Zurich.

Originally published at:



A. Hueni (1), M. Kneubuehler (1)

(1) Remote Sensing Laboratories, Institute of Geography, University of Zurich, Switzerland



Field spectroscopy is a fundamental to Cal/Val as it provides a baseline for satellite and airborne measurements. The considerable time and money spent on the collection of accurate and valuable spectral ground reference data calls for a storage approach that maximises the utilisation of these data by data sharing while ensuring long-term usability.

In this paper we present the state of the art of spectral databases on the example of the SPECCHIO database, its application to Cal/Val activities within the framework of the APEX (Airborne Prism EXperiment) project and general conclusions for future capabilities of spectral repositories.


In response to global change, Earth System Sciences strive to gain a holistic understanding of our planet in order to refine Earth System models and in turn utilize them to create relevant information for informed decision-making. The accurate description of the current state of natural systems is of great importance, as such information is used to both parameterise models as well as to define or refine models [1]. Data accuracy has been shown to be a critical factor in Earth system modelling, especially for climate studies where the signals we are trying to detect are very small [2]. Remote sensing technologies are capable of delivering data for the subsequent generation of many of the ECVs (Essential Climate Variables) defined by GCOS [2, 3]. Within the domain of remote sensing, imaging spectrometers are a tool of choice for the accurate mapping of parts of the Earth System [4]. The generation of accurate quantitative information from imaging spectrometer data requires careful instrument laboratory calibration against references, traceable to international standards as well as Cal/Val activities and supporting operations such as rigorous data collection and storage. Cal/Val activities are crucial to the successful application of remote sensing in general and imaging spectroscopy in particular within Earth System Sciences. According efforts are coordinated by the CEOS (Committee on Earth Observation Satellites) Cal/Val working groups [5] and governed by principles established by the Quality Assurance Framework for Earth Observation (QA4EO) in view of the establishment of GEOSS (Global Earth Observing System of Systems) [6].

In this paper we introduce the concept of spectral databases as structural and functional component in the technical landscape of remote sensing infrastructure, demonstrate the application of the spectral database SPECCHIO for Cal/Val activities within the framework of APEX (Airborne Prism EXperiment) flight campaigns and provide suggestions for future capabilities of spectral databases.

2. THE SPECTRAL DATABASE SPECCHIO 2.1. Provenance and Principles

Spectral databases are systems for the organised storage of spectral signatures accompanied by associated metadata [7] and are therefore proposed as the preferred storage component for spectral in-situ data. The spectral database SPECCHIO was developed in the context of the APEX (Airborne Prism EXperiment) project and supports Cal/Val activities involving a multitude of partners by allowing data sharing among these research groups [7, 8].

Spectral databases are not only mere data repositories; they are information systems, answering questions posed by querying agents by providing adequate information. The transformation from signal to data to information is achieved by (a) collection of data in the field or laboratory utilizing various technical contrivances, notably spectroradiometers for the acquisition of the primary data and (b) organizing the data and entering of metadata, i.e. the process of giving structure and meaning to data, consequently transforming it into information [9].

Figure 1. Signals, Data and Information hierarchy (adapted from Chaffey and Wood [10])

_____________________________________________________ Proc. ‘Hyperspectral 2010 Workshop’, Frascati, Italy, 17–19 March 2010 (ESA SP-683, May 2010)


The according hierarchy is visualized in Figure 1, showing the increase in meaning and value, which are effected by sensing, sampling, data structuring, processing and relating with metadata. The hierarchy is based on the popular DIKW (Data-Information-Knowledge-Wisdom) concept [11], but has, for our purpose, been modified by adding a signals level [12] and omitting both knowledge and wisdom. The current state of the art of spectral databases does not justify adding the latter two tiers, whose inclusion will require further research and development.

It is of interest to note that, for data to qualify as information according to RSDI (Revised Standard Definition of Information), they must be (a) a non-empty set, (b) well formed, (c) meaningful and (d) truthful [13]. While information is required to be truthful, as otherwise we would speak of misinformation, it also important to register that an accuracy (i.e. uncertainty) and precision can be associated with information, defining the adequacy with which the information can answer a question posed by an agent [14]. In the case of a spectral database, it is therefore required to store data in a well structured and meaningful (e.g. described by metadata) and truthful (e.g. removal of wrong measurements) manner with associated uncertainties allowing judgement whether responses of the system are adequate to the accuracy required by a posed problem.

2.2. Components and Functionality

The SPECCHIO system is based upon Java, MySQL and PHP, allowing the operation in heterogeneous computing environments. It is comprised of three main parts: (a) a database hosted on a database server, which holds the multiuser enabled SPECCHIO schema, (b) a Java client application, which allows the interaction with the database using graphical user interfaces and (c) a SPECCHIO web site consisting of a collection of PHP scripts, used for online account creation and distribution of documents and software. Figure 2 illustrates the technical setup of the SPECCHIO system on the example of the publicly accessible SPECCHIO Online System.

The SPECCHIO Java application provides database interfaces for two major functions: (a) Data Ingestion and Management, which includes data loading routines optimized for automated metadata retrieval, metadata editing and instrumentation administration and (b) Data Retrieval and Processing, offering interfaces for query building, data reports, user configurable processing chains and file exports.

A key software design aspect was the minimization of manual user input. For this reason, a number of native spectroradiometer file formats are supported, as this allows the automated extraction of metadata from the input files. In the case of ASD binary files [15], about half of the 34 standard metadata parameters can be

filled automatically. Currently supported input file formats include: ASD, GER, Apogee, ENVI Spectral Library and Formatted Text File.

Figure 2. Technical setup of SPECCHIO Online System

Data sharing and long-term data usage are facilitated by a comprehensive metadata model, which currently comprises 34 standard and an unrestricted number of user-definable numeric parameters.


SPECCHIO is an integral part of APEX operations, storing spectral ground reference data acquired during field campaigns simultaneously to APEX imaging flights. It allows central target signatures access to all partners participating in APEX flight campaigns and thus aids the calibration, validation and interpretation of APEX products. During the 2009 campaigns, a total of over 3000 spectra were collected, entered into the database and augmented with metadata (see Figure 4 for an example of a data report).

Figure 3. Mapping of sampling locations using ArcMap

The recording of spatial positions combined with the MySQL ODBC interface allows direct connection with geographic information systems, such as ArcGIS, for the mapping of spectral ground reference points (Figure 3).


Figure 4. Data report for spectra taken from a tennis court

Figure 5. Change in irradiance at 547nm during 1h30’, documented by reference panel readings (Wm-2sr-1nm-1)

acquired on a day with some cirrus clouds and pointing to the fact that a more frequent acquisition of panel measurements might have been advisable.

With respect to Cal/Val, the following information can be delivered by the SPECCHIO system:

! Radiance data as acquired in the field

! Reflectance data compensated for Spectralon [16] panel deficiencies

! Spatial position and sampling time

! Sun angles, automatically calculated from sampling time and spatial position

! Observation angles (automatically calculated for FIGOS (Field Goniometer System) [17])

! Information about capturing sensor and associated Spectralon Panel (calibration history & uncertainty)

! Irradiance change over time, extracted from reference panel readings (Figure 5)

! Pictures of the sampled ground targets ! Landcover type according to CORINE [18] ! Target type and species name

! Categorical or quantitative description of atmospheric and environmental conditions ! Additional numeric parameters of further

measurements taken from the object, e.g. water content


The utilisation of spectral databases for the organised storage of spectral ground reference in support of Cal/Val is a move towards better quality, cost and work efficiency in the long run. However, it cannot be denied that there is still a huge potential to be explored and exploited, namely:

! Increased support for metadata capturing (e.g. automated loading from spreadsheets).

! Intelligent algorithms for semi-supervised

outlier detection (i.e. controlling the

truthfulness of the data).

! Enhancement of the data model to allow the storage of value added information, i.e. information produced from either data or other information such as reflectance factor

calculation from radiance level or

target/sampling site statistics with a associated provenance records, i.e. fully transparent processing history and trace-back to original data.

! Full uncertainty propagation through the whole chain from signals to information.

! Automated generation of quality indicators, i.e. so-called derived data, allowing suitability analysis of existing data to new applications. ! Specialised interfaces to common remote

sensing software such as ENVI [19], BEAM [20] or Modtran [21] for increased productivity and seamless, automatic integration of Cal/Val and simulation algorithms.

! Integration as component into imaging

spectrometer processing and archiving

facilities for Cal/Val, simulation and algorithm parameterisation as outlined in Hueni et al. [22]

! Integration as a component into complete observing systems such as GEOSS [23].



Cal/Val is an important aspect in the generation of high accuracy products from remote sensing systems [2]. The collection of spectral in-situ data by field spectroscopy provides a baseline for satellite and airborne measurement [24] and is consequently an indispensible part of Cal/Val. The launch of new spaceborne imaging spectrometers such as EnMap [25] and PRISMA [26] will certainly give rise to specific Cal/Val experiments involving spectral ground reference data. Consequently, the adoption of structured storage approaches for these

in-situ data would benefit both data sharing and

long-term usability with positive impacts on product quality assessments.

The true value of spectral data collections can only be fully exploited if data sharing within and across research groups is advanced. To reach these goals, it is required to adhere to rigorous sampling protocols, carry out meticulous primary data and metadata collections and enter these data in spectral databases, consequently enabling other users to search and retrieve valuable and adequate information [27].

In pursuit of this vision and to provide a test environment to potential users, an online version of the SPECCHIO system is provided by the Remote Sensing Laboratories, University of Zurich. To get a personal

account, please browse to


[1] P. Cox and D. Stephenson, "A Changing Climate for Prediction," Science, vol. 317, pp. 207-208, 2007.

[2] S. Ward, "The Earth Observation Handbook - Climate Change Special Edition 2008," ESTEC, Noordwijk, The Netherlands, 2008. [3] GCOS, "The Second Report on the Adequacy of the

Global Observing Systems for Climate in

Support of the UNFCCC," World

Meteorological Organization (WMO), Geneva, Switzerland, 2003.

[4] M. Schaepman, S. Ustin, A. Plaza, T. H. Painter, J. Verrelst, and S. Liang, "Earth system science

related imaging spectroscopy—An

assessment," Remote Sens. Environ., pp. 123-137, 2009.

[5] CEOS, "Cal/Val Portal." [Online]. Available:

[6] QA4EO, "The Quality Assurance Framework for Earth Observation." [Online]. Available:

[7] A. Hueni, J. Nieke, J. Schopfer, M. Kneubühler, and K. Itten, "The spectral database SPECCHIO for improved long term usability

and data sharing," Computers & Geosciences, vol. 35, pp. 557-565, 2009.

[8] K. Itten, F. Dell'Endice, A. Hueni, M. Kneubühler, D. Schläpfer, D. Odermatt, F. Seidel, S. Huber, J. Schopfer, T. Kellenberger, Y. Bühler, P. D'Odorico, J. Nieke, E. Alberti, and K. Meuleman, "APEX - the Hyperspectral ESA Airborne Prism Experiment," Sensors, vol. 8, Special Issue, 2008.

[9] J. Rowley, "The wisdom hierarchy: representations of the DIKW hierarchy," Journal of

Information Science, vol. 33, pp. 163–180,


[10] D. Chaffey and S. Wood, Business Information

Management: Improving Performance Using Information Systems. Harlow: FT Prentice

Hall, 2005.

[11] R. L. Ackoff, "From data to wisdom," Journal of

Applied Systems Analysis, vol. 16, pp. 3–9,


[12] C. W. Choo, "The knowing organization: how organizations use information to construct meaning, create knowledge and make

decisions," International Journal of

Information Management, vol. 16, pp. 329–

340, 1996.

[13] L. Floridi, "Is Information Meaningful Data?,"

Philosophy and Phenomenological Research,

vol. 70, pp. 351-370, 2005.

[14] L. Floridi, "Understanding Epistemic Relevance,"

Erkenntnis, vol. 69, pp. 69-92, 2008.

[15] Analytical Spectral Devices Inc., "Technical Guide," Analytical Spectral Devices Inc., Boulder, CO 20. Mar., 2007.

[16] Labsphere, "Spectralon Care and Handling Guidelines," Labsphere, North Sutton, NH 03260.

[17] J. Schopfer, S. Dangel, M. Kneubühler, and K. I. Itten, "The Improved Dual-view Field Goniometer System FIGOS," Sensors, vol. 8, pp. 5120-5140, 2008.

[18] European Commission DG XI, "CORINE land cover," European Commission Directorate-General Environment, Nuclear Safety and Civil Protection, Office for Official Publications of the European Communities, Luxembourg, 1993.

[19] ITT, "ENVI," 2008

[20] Brockmann Consulting, "BEAM," 2010

[21] A. Berk, L. S. Bernstein, and D. C. Robertson, "MODTRAN: A Moderate Resolution Model for LOWTRAN7," Air Force Geophys. Lab., Bedford, MA, Report GL-TR-89-0122, 1989. [22] A. Hueni, J. Biesemans, K. Meuleman, F.

Dell'Endice, D. Schläpfer, S. Adriaensen, S. Kempenaers, D. Odermatt, M. Kneubuehler, J. Nieke, and K. Itten, "Structure, Components


and Interfaces of the Airborne Prism Experiment (APEX) Processing and Archiving Facility," IEEE Transactions on Geoscience

and Remote Sensing, vol. 47, pp. 29-43, 2009.

[23] GEO, Global earth observation system of systems

GEOSS: 10-Year implementation plan reference document. 2200 AG Noordwijk, The

Netherlands ESA Publications Division 2005. [24] GCOS, "Guideline for the Generation of

Satellite-based Datasets and Products meeting GCOS

Requirements," World Meteorological

Organization (WMO), Geneva, Switzerland, 2009.

[25] T. Stuffler, K. Forster, S. Hofer, M. Leipold, B. Sang, H. Kaufmann, B. Penne, A. Mueller, and C. Chlebek, "Hyperspectral imaging-An advanced instrument concept for the EnMAP mission (Environmental Mapping and Analysis Programme)," Acta Astronautica, vol. 65, pp. 1107-1112, Oct-Nov 2009.

[26] D. Labate, M. Ceccherini, A. Cisbani, V. De Cosmo, C. Galeazzi, L. Giunti, M. Melozzi, S. Pieraccini, and M. Stagi, "The PRISMA payload optomechanical design, a high

performance instrument for a new

hyperspectral mission," Acta Astronautica, vol. 65, pp. 1429-1436, Nov-Dec 2009.

[27] K. Pfitzner, A. Bollhöfer, and G. Carr, "Standards

for reflectance spectral measurement,"

Australian Government, Department of the Environment, Water, Heritage and Arts, Supervising Scientist, Darwin, NT, Australia, 2009.





Related subjects :