Teseo: A vectoriser of historical seismograms

doi:10.1016/j.cageo.2005.04.001

Computers & Geosciences

Volume 31, Issue 10, December 2005, Pages 1277-1285

https://doi.org/10.1016/j.cageo.2005.04.001 Get rights and content

Abstract

Historical seismograms contain a rich harvest of information useful for the study of past earthquakes. It is necessary to extract this information by digitising the analogue records if modern analysis is required. Teseo has been developed for quick and accurate digitisation of seismogram traces from raster files, introducing a vectorisation step based on piecewise cubic Bézier curves. The vectoriser can handle greyscale images stored in a suitable file format and it offers three concurrent vectorisation methods: manual, automatic by colour selection, and automatic by neural networks. The software that implements the methods described is distributed with open source license.

Introduction

The use of modern techniques to recover seismological information contained in historical seismograms can supply additional knowledge on past seismicity and ongoing tectonic processes. This important goal is widely acknowledged (Kanamori, 1988, Stein et al., 1988), but it presents some difficulties, since data recorded by early instruments on paper media must be properly processed to obtain numerical data usable for modern analysis. The Sismos Project¹ is aimed at preserving the heritage of historical seismograms owned by all the Italian observatories. For this purpose, an acquisition of the raster images from approximately one million recordings is planned. Data access will be ensured through a dedicated web portal. Digitisation will be performed on a number of the most important events. All this requires a fast and accurate digitisation procedure that is usable on images of large dimensions and independent of the original paper type or recording instrument.

In the digitisation process, seismograms can be classified based on the similarity of the raster images used. Common problems encountered in automatic digitisation of seismic traces are well described by Trifunac et al. (1999). Other recent work on seismogram digitisation includes that by Samardjieva et al. (1998), who created a digital database for historical earthquakes using a manual digitisation process whereby the original records were enlarged by a projector on a screen. Teves-Costa et al. (1999) presented an example of the recovery of source parameters from historical records and developed a semi-automatic method using commercial software on images of 200 dpi resolution. Baskoutas et al. (2000) digitised 1852 seismograms obtained from the Mainka and Wiechert seismographs in the National Observatory of Athens during the period 1911–1960. They developed software usable on black and white images with 1600, or 500 dpi if the image was too large.

The digitisation procedure usually involves:

(i)
extraction of the sample sequence directly from the image, in a manual or automatic way,
(ii)
correct mapping from the $(x, y)$ image coordinates to the amplitude and time of the samples.

We present here a different method that relies on an intermediate parametric vectorial representation of the seismogram trace using piecewise cubic Bézier curves. To implement this method, a software tool named Teseo has been developed. It offers one manual and two automatic modes of operation. The first automatic mode is analytical and the second is based on neural networks.

Section snippets

Data

Analogue seismograms recorded on paper result from the response of a seismometer and a recording system to ground motion. The main classes of traditional seismometers include short-, intermediate- and long-period instruments. The recording system for analogue instruments is characterised by several mechanisms, the most important being the kind of support used (smoked, photographic, thermal), the type of tracing device (needle, light beam) and the paper speed. The digitisation of traces

Teseo vectoriser

The standard seismogram digitisation process is shown in a flow chart in Fig. 5. The Image process produces a pixel coordinate sequence for the trace (rough digital) and the Correction process transforms it in a time domain to a seismic signal (final digital). This latter stage needs some instrumental parameters, either known a priori or obtainable from the seismogram image.

Within the Image process, we have introduced an intermediate step that produces a vectorial representation of the seismic

Vectorising with Teseo

In this section, some examples of vectorisation with the various automatic systems illustrated here are shown. Fig. 10(A) shows an example of automatic vectorisation of a smoked paper seismogram image. This image presents little curvature distortion and some crossings. Vectorisation was successfully obtained using the weighted mean colour algorithm. Fig. 10(B) shows the vectorisation during the first stage, in which many Bézier segments were found. The resulting Bézier path after least-squares

Comments and conclusions

A vectorisation method for seismic records is presented, together with the software used. The vectorisation output is a piecewise cubic Bézier curve that can be resampled. A feature of Teseo is that it adopts both higher dpi images and long greyscale depth. Two characteristics distinguish this system from previous work. A manual and two semi-automatic vectorisation methods have been described. Manual vectorisation is accurate but time-consuming. The colour trace algorithm is well suited in many

Acknowledgements

The authors would like to thank A. Michelini for his useful and constructive suggestions and for his continuous encouragement. They are also grateful to B. De Simoni for giving them the opportunity to start this work. Further thanks are addressed to B. Palombo, who uses Teseo for her studies, and to L. Arcoraci, who was a dedicated beta tester.

References (28)

L. Atzori et al.
Reconstruction of missing or occluded contour segments using Beziér interpolations
Signal Processing
(2000)
S. Baek et al.
A robot endeffector tracking system based on feedforward neural networks
Robotics and Autonomous Systems
(1999)
P.E. Bézier
Mathematical and practical possibilities of unisurf
H. Chang et al.
Vectorization of hand-drawn image using piecewise cubic Bézier curves fitting
Pattern Recognition
(1998)
L. Cinque et al.
Shape description using cubic polynomial Beziér curves
Pattern Recognition Letters
(1998)
M. Egmont-Petersen et al.
Image processing with neural networks-a review
Pattern Recognition
(2002)
P. Goldstein et al.
SAC2000signal processing and analysis tools for seismologists and engineers
L. Shao et al.
Curve fitting with Bézier cubics
Graphical Models and Image Processing
(1996)
M.D. Trifunac et al.
Common problems in automatic digitization of strong motion accelerograms
Soil Dynamics and Earthquake Engineering
(1999)
Adobe Developers Association, 1992. TIFF 6.0. Adobe Systems Incorporated—1585 Charleston Road—P.O. Box 7900—Mountain...

Autodesk, 1992. AutoCAD Release 12 DXF Format. Adobe Systems Incorporated—1585 Charleston Road—P.O. Box 7900—Mountain...

I.G. Baskoutas et al.

A modern technique for the retrieval and processing of historical seismograms in Greece

Natural Hazards

(2000)

C. Bunks

Grokking the GIMP

(2000)

M. Huang et al.

The pre-processing of data points for curve fitting in reverse engineering

International Journal of Advanced Manufacturing Technology

(2000)

Cited by (44)

Source parameters of the 1926 and 1927 Jersey earthquakes from historical, instrumental, and macroseismic data
2020, Physics of the Earth and Planetary Interiors
Citation Excerpt :
Seismograms were scanned by the staff of the SISMOS Project of the Istituto Nazionale di Geofisica e Vulcanologia (INGV) in Rome. Then, traces were digitized manually through a methodology developed by SISMOS using the Teseo2 software (Pintore et al., 2005); details on SISMOS and the Gimp plug-in Teseo2 are available atteseo.rm.ingv.it (last accessed in January 2019). The manual digitization of seismograms from scanned images is the most reliable option and is the most widespread practice to date (Bent, 1992;Baroux et al., 2003;Stich et al., 2005;Dost and Haak, 2006;Batlló et al., 2010).
English Channel Islands are located off Normandy coast of France within an intraplate area not associated with high seismicity rate and active tectonics. However, in July 1926 a damaging and well-documented earthquake occurred there, followed by a strongly felt event in February 1927. In this paper, we reprocess macroseismic observations, analog seismograms and bulletin data in order to re-appraise the location and magnitude of these two earthquakes. We find that the macroseismic epicentre of the 1926 Jersey earthquake is shifted to the East when compared with the location offshore South of Jersey given in both the French database SisFrance and the recent French catalog FCAT-17. Arrival-times from published data, together with our own onset-time readings, are processed in order to obtain probabilistic hypocentral locations. The maximum-likelihood instrumental epicentre of the 1926 event is located about 15 km East of Jersey Island (49.20°N, 1.82°W). This epicentre is well constrained within a 10 km radius area. The location of the 1927 epicentre is slightly south of Jersey, at a location similar to that of SisFrance, but this latter epicentre is less constrained as fewer observations are available (the Probability Density Function for the 1927 epicentre is relatively “diluted”). Focal depth remains very poorly determined for both the 1926 and 1927 events. Analysis of historical seismograms is also performed in order to determine both surface-wave and moment magnitudes, M_S and M_W. We find M_S = 5.6 ± 0.2 for the 1926 event and M_S = 5.0 for the 1927 event. Waveform fitting of some of the highest quality seismograms of the 1926 event gives us M_W in the range [5.0, 5.5], depending on the chosen focal depth and focal mechanism. The 1927 Jersey earthquake is expected with an intensity magnitude MI about 0.6 smaller.
A reappraisal of the 1978 Ferruzzano earthquake (southern Italy) from new estimates of hypocenter location and moment tensor inversion
2019, Physics of the Earth and Planetary Interiors
Citation Excerpt :
Each seismogram has been digitized manually with the software GIMP (GNU Image Manipulation Program) by redrawing the whole traces. Then it has been interpolated and converted to modern seismological format (i.e., SAC) by TESEO2 (Turn the Eldest Seismograms into the Electronic Original ones) a GIMP plug-in made up by INGV (Pintore et al., 2005). Finally, all waveforms have been corrected for geometrical distortions and first arrivals have been accurately verified.
The March 11th, 1978 Ferruzzano earthquake is the most recent moderate-to-major earthquake occurred in the southern Calabrian region (southern Italy), one among the highest seismic risk areas of the whole Mediterranean. Previous information available from the literature on the 1978 earthquake is quite contrasting and not well framed in the regional seismotectonic scenario. In the present study we selected and digitized analog seismograms coming from stations of the Euro-Mediterranean region to invert for the deviatoric seismic moment tensor through a time-domain algorithm properly implemented to analyze data recorded before the advent of the digital era. Moreover, we estimated a new hypocentral location by using original bulletin data and a non-linear probabilistic earthquake location technique working with 3D velocity models. The quality and stability of the obtained results, both for hypocenter location and moment tensor inversion, were accurately checked by several inversion tests. Our results indicate that the 1978 earthquake (i) occurred westward and at a shallower depth respect to previous hypocenter locations, (ii) is characterized by a ca. N-S trending normal faulting mechanism and (iii) has a moment magnitude of 4.7, thus suggesting an overestimate of previous evaluations. This study furnishes new information on the 1978 Ferruzzano earthquake allowing to better frame it in the regional seismotectonic scenario and also proves that the time-domain waveform inversion algorithm applied to digitized old seismograms is capable to successfully invert also M_w < 5 earthquakes. The obtained results pave the way for future analyses of the early instrumental seismicity potentially capable to furnish new constraints to local and regional seismotectonic modeling.
A new program on digitizing analog seismograms
2016, Computers and Geosciences
Historical seismograms contain a great variety of useful information which can be used in the study of earthquakes. It is necessary for researchers to digitize analog records and extract the information just as modern computing analysis requires. Firstly, an algorithm based on color scene filed method is presented in order to digitize analog seismograms. Secondly, an interactive software program using C# has been developed to digitize seismogram traces from raster files quickly and accurately. The program can deal with gray-scale images stored in a suitable file format and it offers two different methods: manual digitization and automatic digitization. The test result of the program shows that the methods presented in this paper can lead to good performance.
Historical seismograms: Preserving an endangered species
2015, GeoResJ
Citation Excerpt :
The database is presently (2014) being reorganized, with the aim of providing access to low resolution (200 bpi) scans directly online, and to the full high resolution scans (600 to 1024 bpi) through customized requests. As part of the SISMOS project, the digitizing software “Teseo” was developed by Pintore et al. [51]. The Silverton (Pretoria) branch of the Council for Geosciences (South Africa) holds a large collection of analog seismograms, including the Milne-Shaw instrument operated at Cape Town between 1920 and 1938.
The youth of seismology as a science, compared to the typical duration of seismic cycles, results in a relative scarcity of records of large earthquakes available for processing by modern analytical techniques, which in turn makes archived datasets of historical seismograms extremely valuable in order to enhance our understanding of the occurrence of large, destructive earthquakes. Unfortunately, the value of these datasets is not always perceived adequately by decision-making administrators, which has resulted in the destruction (or last-minute salvage) of irreplaceable datasets.
We present a quick review of the nature of the datasets of seismological archives, and of specific algorithms allowing their use for the modern retrieval of the source characteristics of the relevant earthquakes. We then describe protocols for the transfer of analog datasets to digital support, including by contact-less photography when the poor physical state of the records prevents the use of mechanical scanners.
Finally, we give some worldwide examples of existing collections, and of successful programs of digital archiving of these valuable datasets.
Digitizing analog seismograms of Iranian seismic stations — preserving records for strong earthquakes in Iran 1960–1990
2023, Journal of Seismology
A refined catalogue of focal mechanisms for the Intra-Carpathian region of Romania: implications for the stress field and seismogenic features assessment
2023, Annals of Geophysics

View all citing articles on Scopus

^☆: Source code is available from server at http://sismos.ingv.it

View full text

Teseo: A vectoriser of historical seismograms☆

Abstract

Introduction

Section snippets

Data

Teseo vectoriser

Vectorising with Teseo

Comments and conclusions

Acknowledgements

Signal Processing

Robotics and Autonomous Systems

Pattern Recognition

Pattern Recognition Letters

Pattern Recognition

Graphical Models and Image Processing

Soil Dynamics and Earthquake Engineering

A modern technique for the retrieval and processing of historical seismograms in Greece

Natural Hazards

Grokking the GIMP

The pre-processing of data points for curve fitting in reverse engineering

International Journal of Advanced Manufacturing Technology