magistrsko delo
Abstract
V tem delu je opisan postopek za določitev zunanjih koordinat objekta z uporabo dveh kamer in pomočjo metod strojnega vida. Sistem na ta način omogoča brezstično določitev zunanjih koordinat objekta oz. aktivnega markerja. Izdelali smo dva markerja in preizkusili njuno točnost določanja zunanjih koordinat. Pridobljeno znanje iz prvega sklopa dela smo uporabili še za določanje zunanjih koordinat orientacije glave. V začetnih poglavjih so opisane osnove digitalne fotografije in teoretično ozadje popačitev, ki so prisotne ob zajetju slike s kamero. Opisan je tudi postopek modeliranja omenjenih popačitev. Naslednja poglavja zajemajo pristop in načrtovanje sistema z markerjem. Pri tem je največji izziv predstavljala ocena točnosti pozicije in ocena hitrosti sledenja markerja. V nadaljevanju je opisana še metoda, izračuna zunanjih koordinat orientacije glave, ter problemi, ki so se pojavili med samim delom. Na koncu so podani merilni rezultati primerjav obeh markerjev, ter uspešnost določanja pozicije glave.
Keywords
določanje pozicije objekta;kalibracija kamere;določanje orientacije glave;OpenCV;
Data
Language: |
Slovenian |
Year of publishing: |
2015 |
Typology: |
2.09 - Master's Thesis |
Organization: |
UM FERI - Faculty of Electrical Engineering and Computer Science |
Publisher: |
D. Šipoš |
UDC: |
004.896:681.5(043.2) |
COBISS: |
19401750
|
Views: |
1039 |
Downloads: |
196 |
Average score: |
0 (0 votes) |
Metadata: |
|
Other data
Secondary language: |
English |
Secondary title: |
ESTIMATION OF OBJECT`S EXTERNAL COORDINATES USING TWO CAMERAS |
Secondary abstract: |
This master’s thesis presents an estimation of objects external coordinates using two cameras. The system in this way allows us contactless determination of external object coordinates. We present to use an evaluate accuracy estimate of the external coordinates. The knowledge that we have gained from the first system we also used to estimate the external orientation coordinates of human head. Beginning chapters describe the basics of digital photography and theoretical background of lens distortion, when an image was acquired with a camera. Outlined is also the procedure for modeling the lens distortion. The sections covers the design of a marker. The biggest challenge in this approach was to achieve accuracy of markers position and its estimation within 20 ms. Next sections describe the method for estimation the external coordinates of head orientation, and the problems that have arisen during the thesis. The last chapters presents measurement results. Conclusion concludes the thesis. |
Secondary keywords: |
object pose estimation;camera calibration;head pose estimation;OpenCV; |
URN: |
URN:SI:UM: |
Type (COBISS): |
Master's thesis/paper |
Thesis comment: |
Univ. v Mariboru, Fak. za elektrotehniko, računalništvo in informatiko, Elektrotehnika |
Pages: |
IX, 60 f. |
ID: |
9078974 |