Maik Veste1, Gerhard Kast2, Hans-Peter Schäfer3, Rüdiger Harms4
1 CEBra - Centrum für Energietechnologie Brandeburg e.V., Friedlieb-Runge-Strasse 3, 03046 Cottbus
Email: maik.veste@me.com
2 UP Umweltanalytische Produkte GmbH, Taubenstrasse 4, 03046 Cottbus
3 Schäfer GmbH - Industrielle Meß- und Videotechnik, Trautenstrasse 6, 38118 Braunschweig
4 PhysComp, Sperlingsgasse 17A, 38126 Braunschweig
Zusammenfassung: Ein vollautomatisches Minirhizotronsystem wurde in den vergan-genen Jahren entwickelt und erfolgreich getestet. Das System basiert auf einer Finger-kamera und erlaubt ein vollautomatisches Erfassen von Wurzeln bis zu einer Bodentiefe von 160 cm.
Deskriptoren: Automatisches Minirhizotronsystem, Wurzelsystem, Fingerkamera
Abstact: A fully automatic minirhizotron system was developed in the recent years and successfully tested. The minirhiztron system is based on a finger cam (Sony) and allows a fully automatic detection of roots until a soil depth of 160 cm.
Keywords: automated minirhizotron system, root system, finger camera
1 Introduction
The root systems of plants play an important part in the carbon allocation and for the uptake of nutrients and water. However, research on plant root systems under field conditions is difficult, because the soil limits the direct observation. Soil and ingrowth cores give only estimates of standing root biomass and relative growth and they are destructive. On the other hand, the minirhizotron method permit the measurement of fine root production, mortality and turn-over (MAJDI et al. 2005, VESTE 2012). Absolute values of fine root production and mortality can be estimated by combining data from minirhizotrons and soil cores. Seasonal changes in the root dynamic can be related directly to above ground production. Minirhizotrons can provide qualitative information on root color, branching and the development of mycorrhiza. The minirhizotron technique can be used to monitor the same root(s) over selected time intervals, which
Bornimer Agrartechnische Berichte • Heft 88 ISSN 0947-7314
Leibniz-Institut für Agrartechnik Potsdam-Bornim e.V. (ATB)
21. Workshop Computer-Bildanalyse in der Landwirtschaft und
3. Workshop Unbemannte autonom fliegende Systeme in der Landwirtschaft 139
can vary from days to years. In the last two decades different techniques were developed for the in-situ root observations: (i) video technique (FERGUSON & SMUCKER
1989, AMATO et al. 2012), (ii) endoscopes (VESTE 2012) and (iii) scanner-systems (DANNOURA et al. 2008). We developed a fully automatic minirhizotron system (Figure1, VESTE et al. 2013) to optimize the workflow and to improved the accuracy of the inventory of the root systems.
Figure 1: Computer-controlled automatic minirhizotron system.
2 Automatic minirhizotron system
The minirhiztron system is based on a finger cam (Sony) and allows a fully automatic scan of roots in the soil. For the root recording a color and a monochrome camera with a diameter of 28 mm were developed. The cameras use a PAL signal, which is converted directly by a USB-video converter into a digital image (JPEG). In order to minimize the reflections in the glass tubes indirect illumination of LED are used. The light is located in the lower part of the camera. Computer-controlled step motors allows x-/y-positioning of the camera along the glass tube with a high and reliable accuracy.
For the positioning of the camera three different operational mode are available:
manual movement in the tube;
computer-controlled positioning of the camera at elected coordinates (depth, angle);
fully automatic scan of the entire tube in freely definable depth ranges.
Bornimer Agrartechnische Berichte • Heft 88 ISSN 0947-7314
Leibniz-Institut für Agrartechnik Potsdam-Bornim e.V. (ATB)
Veste, Kast, Schäfer, Harms 140
The recorded picture is stored on the hard discs in JPEG format as individual image and assembled to one large image of the entire tube. Examples of root captures from the black and white camera and of the color camera are shown in Figure 2.
Figure 2: (A) Root image of the black and white camera and (B) of the color camera.
The special software ROOTS optimized the image quality, reduce noises and enhance the contrast of the image (HARMS et al. 2014) for further root detection and analysis with the open-source software RootFly. The future development of the minirhizotrons needs an improved system of image analysis to optimize the data analysis. The high number of root data requires an automatic detection of the roots in the image, but such a system is still under development (ERZ et al. 2005, VESTE 2010).
Acknowledgements
The system was developed by funding provided by Ministerium für Wirtschaft und Eu-ropaangelegenheiten, Brandenburg, Germany.
References
AMATO M.,LUPO F.,BITTELA G.,BOCHICCHIO R.,ABEL AZIZ M,CELANO G. (2012): A high quality low-costs digital microscope minirhizotron system. Computers and Electronics in Agriculture 80: 50-53.
DANNOURA M., KOMINAMI Y., OGUMA H., KANAZAWA Y. (2008): The development of an optical scanner method for observation of plant root dynamics. Plant Root 2: 14-18.
ERZ G.,VESTE M., ANLAUF H., BRECKLE S.-W., POSCH S. (2005): A region and contour based technique for automatic detection of roots of tomatoes in minirhizotron images. Applied Botany and Food Quality 79: 83-88.
FERGUSON J.C., SMUCKER A.J.M. (1989): Modifications of the minirhizotron video camera system for measuring spatial and temporal root dynamics. Soil Sci. Soc. Am. J. 53: 1601–
1605.
Bornimer Agrartechnische Berichte • Heft 88 ISSN 0947-7314
Leibniz-Institut für Agrartechnik Potsdam-Bornim e.V. (ATB)
21. Workshop Computer-Bildanalyse in der Landwirtschaft und
3. Workshop Unbemannte autonom fliegende Systeme in der Landwirtschaft 141
HARMS R., SCHÄFER H.-P., KAST G., VESTE M. (2014): Optimierung der Bildauswertung von Farbbildern aus Minirhizotronen zur Wurzelbeobachtung. Bornimer Agrartechnische Berichte: in diesem Band.
MAJDI H., PREGITZER K., MORÉN A.-S., NYLUND J.-E., ÅGREN G.I.(2005): Measuring fine root turnover in forest ecosystems. Plant and Soils 276: 32-75.
VESTE M.(2010): Anwendung automatischer Verfahren zur Erfassung von Wurzeln in Minirhizot-ron-Bildern. Bornimer Agrartechnische Berichte 73: 72-78
VESTE M. (2012): Anwendung von optischen Messsystemen zur nicht-destruktiven Erfassung von Wurzeln im Boden. Bornimer Agrartechnische Berichte 78: 81-89
VESTE M.,SCHÄFER H.-P.,HARMS J., KAST G. (2013): Ein neues computergesteuertes Minirhi-zotron-System zur Erfassung von Wurzeln. Bornimer Agrartechnische Berichte 81: 295-301.
Bornimer Agrartechnische Berichte • Heft 88 ISSN 0947-7314
Leibniz-Institut für Agrartechnik Potsdam-Bornim e.V. (ATB)
Gieselmann 142