DBPapers
DOI: 10.5593/sgem2017H/33/S14.066

FIRST TEST OF LOW-COST SINGLE-BOARD COMPUTER BASED CAMERA FOR CLOSE-RANGE FOREST PHOTOGRAMMETRY

M. Mokros, M. Koren, J. Cernava, M. Zapotocky, J. Tomastik
Thursday 23 November 2017 by Libadmin2017

References: 17th International Multidisciplinary Scientific GeoConference SGEM 2017, www.sgemviennagreen.org, SGEM2017 Vienna GREEN Conference Proceedings, ISBN 978-619-7408-27-0 / ISSN 1314-2704, 27 - 29 November, 2017, Vol. 17, Issue 33, 529-536 pp; DOI: 10.5593/sgem2017H/33/S14.066

ABSTRACT

Close-range photogrammetry (CRP) represents very perspective technology for many fields. It can provide results comparable to terrestrial laser scanning, and the hardware is still much cheaper. Our research group is developing low-cost CRP device based on Raspberry Pi 3 computer. Raspberry Pi 3 is a single-board computer with relatively high computing power and low price. In connection with a touchscreen, Pi Camera module v2.1 (Sony IMX219, 8Mpx) and powerbank it becomes a portable device for imagery. Continuous shooting based image collection (1 image per second) was carried out using regular RGB camera and NoIR (no infrared filter) camera at a test plot. Images of the same area were collected by DSLR camera Canon EOS 70D (20.2 megapixels) as a reference to our low-cost solution. The images were processed into a form of dense point cloud by Pix4D software package, and the diameters were estimated by DendroCloud software (gis.tuzvo.sk/dendrocloud/) using an optimal circle fitting method. On-line visualisation of the point clouds is available at maps.tuzvo.sk/sgem2017/. Diameters at three different heights above ground (0.65m, 1.3m and 2m) were estimated for a sample of four trees. Reference diameters were measured by a calliper. The root mean square error (RMSE) and relative RMSE of diameter estimation for regular RGB were 2.13cm and 6.9%; 2.54cm and 8.3% for NoIR camera. The RMSE and relative RMSE for diameters derived from point cloud created using Canon images were 1.18cm and 3.8% respectively. Results of our first test showed that Raspberry Pi camera solution has a great potential for the use in forest inventory. Future research will focus on testing developed device on larger plots and the development of a multi-camera system based on single-board computers. This work was supported by the Slovak Research and Development Agency under the contract No. DS-2016-0040 (Benchmarking of close-range photogrammetry methods for forestry applications, gis.tuzvo.sk/benchcrp/).

Keywords: Raspberry Pi, close-range photogrammetry, forest, DendroCloud

Home | Contact | Site Map | Site statistics | Visitors : 0 / 353063

Follow site activity en  Follow site activity FOREST ECOSYSTEMS  Follow site activity Papers SGEM2017   ?

CrossRef Member    Indexed in ISI Web Of Knowledge   Indexed in ISI Web Of Knowledge
   

© Copyright 2001 International Multidisciplinary Scientific GeoConference & EXPO SGEM. All Rights Reserved.

Creative Commons License