diplomska naloga
Lidija Dugon (Author), Simona Savšek (Mentor), Gašper Štebe (Co-mentor)

Abstract

Na težko dostopnih mestih, kjer uporaba reflektorjev ni mogoča, so točke pogosto signalizirane z reflektivnimi nalepkami. Za opazovanje premikov takšnih objektov je zahtevana visoka natančnost, ki jo dosežemo z uporabo ustrezne geodetske opreme, ustrezno izbiro metode dela ter ustrezno signalizacijo točk. Zaradi pogoste uporabe reflektivnih nalepk za signalizacijo smo želeli preizkusiti sistem avtomatskega prepoznavanja tarč pri merjenju na reflektivne nalepke. Sistem avtomatskega merjenja smo primerjali z ročnimi meritvami in ugotavljali skladnost meritev. Za izvedbo meritev smo si vzpostavili testno bazo, kjer smo uporabili dva sodobna elektronska instrumenta proizvajalca Leica Geosystems. Merili smo na reflektivne nalepke, ki so bile različnih velikosti, z različnim zasukom in na različni oddaljenosti. Na posamezno reflektivno nalepko smo v večih ponovitvah merili horizontalne smeri, zenitne razdalje in poševne dolžine. Z uporabo statističnih metod smo želeli ugotoviti, ali je pri merjenju na reflektivne nalepke smiselno uporabiti sistem avtomatskega prepoznavanja tarč. Rezultati analize so pokazali, da so avtomatske meritve slabše natančnosti kot ročne. Ročne in avtomatske meritve so v večini primerov statistično različne. Povzamemo lahko, da uporaba avtomatskega sistema (APT) na reflektivne nalepke za izvedbo preciznih meritev ni zanesljiva.

Keywords

diplomska naloga;UNI;GIG;reflektivne nalepke;Leica Geosystems MS50;Leica Geosystems TS30;sistem APT;ocena natančnosti;statistična analiza;

Data

Language: Slovenian
Year of publishing:
Typology: 2.11 - Undergraduate Thesis
Organization: UL FGG - Faculty of Civil and Geodetic Engineering
Publisher: [L. Dugon]
UDC: 528.53(043.2)
COBISS: 8528737 Link will open in a new window
Views: 906
Downloads: 379
Average score: 0 (0 votes)
Metadata: JSON JSON-RDF JSON-LD TURTLE N-TRIPLES XML RDFA MICRODATA DC-XML DC-RDF RDF

Other data

Secondary language: English
Secondary title: Testing automatic target recognition (ATR) system on retro reflective targets
Secondary abstract: In hard-to-reach places, where the use of standard reflectors is not possible, points are often signalized by retro reflective targets. To determine the movements of such objects, high precision is required, which is achieved by using the appropriate geodetic equipment, appropriate selection of the measuring method, and appropriate signalling of points. Due to the frequent use of retro reflective targets, we wanted to test the automatic target recognition system on retro reflective targets. We have compared measurements, obtained by automatic target recognition system with manual measurements. In order to carry out the measurements, we established a test field. Measurements were obtained by two modern electronic instruments manufactured by Leica Geosystems. Retro reflective targets, used as a signal, were different sizes, with a different rotation and at different distances. We measured horizontal directions, zenith distances and slope distance in multiple repetitions on several retro reflective targets. By using statistical methods, we wanted to determine whether it is sensible to use the automatic target recognition system when measuring on retro reflective targets. Based on analysis result we concluded, that automatic measurements are less accurate than manual. Manual and automatic measurements are in most cases statistically different, which can be summarized: automatic measurements on retro reflective targets are not reliable.
Secondary keywords: Retro reflective targets;Leica Geosystems MS50;Leica Geosystems TS30;automatic target recognition;accuracy assessment;statistical analysis;
Type (COBISS): Bachelor thesis/paper
Study programme: 0
Embargo end date (OpenAIRE): 1970-01-01
Thesis comment: Univ. v Ljubljani, Fak. za gradbeništvo in geodezijo
Pages: XII, 47 str., 12 pril.
ID: 10959785